├── .gitignore ├── Dockerfile ├── LICENSE ├── README.md ├── environment.yaml ├── sample_images ├── clean_1000iters.png └── noisy.png └── src ├── cscnet.py ├── example.yaml ├── fmd_dataloader.py ├── inference.py ├── model.py ├── neighbor2neighbor.py ├── run_folder.py └── training.py /.gitignore: -------------------------------------------------------------------------------- 1 | # Byte-compiled / optimized / DLL files 2 | __pycache__/ 3 | *.py[cod] 4 | *$py.class 5 | 6 | # C extensions 7 | *.so 8 | 9 | # Distribution / packaging 10 | .Python 11 | build/ 12 | develop-eggs/ 13 | dist/ 14 | downloads/ 15 | eggs/ 16 | .eggs/ 17 | lib/ 18 | lib64/ 19 | parts/ 20 | sdist/ 21 | var/ 22 | wheels/ 23 | share/python-wheels/ 24 | *.egg-info/ 25 | .installed.cfg 26 | *.egg 27 | MANIFEST 28 | 29 | # PyInstaller 30 | # Usually these files are written by a python script from a template 31 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 32 | *.manifest 33 | *.spec 34 | 35 | # Installer logs 36 | pip-log.txt 37 | pip-delete-this-directory.txt 38 | 39 | # Unit test / coverage reports 40 | htmlcov/ 41 | .tox/ 42 | .nox/ 43 | .coverage 44 | .coverage.* 45 | .cache 46 | nosetests.xml 47 | coverage.xml 48 | *.cover 49 | *.py,cover 50 | .hypothesis/ 51 | .pytest_cache/ 52 | cover/ 53 | 54 | # Translations 55 | *.mo 56 | *.pot 57 | 58 | # Django stuff: 59 | *.log 60 | local_settings.py 61 | db.sqlite3 62 | db.sqlite3-journal 63 | 64 | # Flask stuff: 65 | instance/ 66 | .webassets-cache 67 | 68 | # Scrapy stuff: 69 | .scrapy 70 | 71 | # Sphinx documentation 72 | docs/_build/ 73 | 74 | # PyBuilder 75 | .pybuilder/ 76 | target/ 77 | 78 | # Jupyter Notebook 79 | .ipynb_checkpoints 80 | 81 | # IPython 82 | profile_default/ 83 | ipython_config.py 84 | 85 | # pyenv 86 | # For a library or package, you might want to ignore these files since the code is 87 | # intended to run in multiple environments; otherwise, check them in: 88 | # .python-version 89 | 90 | # pipenv 91 | # According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control. 92 | # However, in case of collaboration, if having platform-specific dependencies or dependencies 93 | # having no cross-platform support, pipenv may install dependencies that don't work, or not 94 | # install all needed dependencies. 95 | #Pipfile.lock 96 | 97 | # poetry 98 | # Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control. 99 | # This is especially recommended for binary packages to ensure reproducibility, and is more 100 | # commonly ignored for libraries. 101 | # https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control 102 | #poetry.lock 103 | 104 | # pdm 105 | # Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control. 106 | #pdm.lock 107 | # pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it 108 | # in version control. 109 | # https://pdm.fming.dev/#use-with-ide 110 | .pdm.toml 111 | 112 | # PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm 113 | __pypackages__/ 114 | 115 | # Celery stuff 116 | celerybeat-schedule 117 | celerybeat.pid 118 | 119 | # SageMath parsed files 120 | *.sage.py 121 | 122 | # Environments 123 | .env 124 | .venv 125 | env/ 126 | venv/ 127 | ENV/ 128 | env.bak/ 129 | venv.bak/ 130 | 131 | # Spyder project settings 132 | .spyderproject 133 | .spyproject 134 | 135 | # Rope project settings 136 | .ropeproject 137 | 138 | # mkdocs documentation 139 | /site 140 | 141 | # mypy 142 | .mypy_cache/ 143 | .dmypy.json 144 | dmypy.json 145 | 146 | # Pyre type checker 147 | .pyre/ 148 | 149 | # pytype static type analyzer 150 | .pytype/ 151 | 152 | # Cython debug symbols 153 | cython_debug/ 154 | 155 | # PyCharm 156 | # JetBrains specific template is maintained in a separate JetBrains.gitignore that can 157 | # be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore 158 | # and can be added to the global gitignore or merged into this file. For a more nuclear 159 | # option (not recommended) you can uncomment the following to ignore the entire idea folder. 160 | #.idea/ 161 | data/ 162 | results/ -------------------------------------------------------------------------------- /Dockerfile: -------------------------------------------------------------------------------- 1 | FROM continuumio/anaconda3 2 | 3 | # Install system dependencies 4 | RUN apt-get update \ 5 | && DEBIAN_FRONTEND=noninteractive apt-get install -y \ 6 | build-essential \ 7 | curl \ 8 | git \ 9 | && apt-get clean 10 | 11 | # Install python miniconda3 + requirements 12 | ENV MINICONDA_HOME="/opt/miniconda" 13 | ENV PATH="${MINICONDA_HOME}/bin:${PATH}" 14 | COPY environment.yaml environment.yaml 15 | RUN conda env create --name p2s --file=environment.yaml 16 | 17 | # 18 | WORKDIR /Poisson2Sparse 19 | COPY src /Poisson2Sparse 20 | 21 | ENTRYPOINT [ "bash"] 22 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | GNU GENERAL PUBLIC LICENSE 2 | Version 3, 29 June 2007 3 | 4 | Copyright (C) 2007 Free Software Foundation, Inc. 5 | Everyone is permitted to copy and distribute verbatim copies 6 | of this license document, but changing it is not allowed. 7 | 8 | Preamble 9 | 10 | The GNU General Public License is a free, copyleft license for 11 | software and other kinds of works. 12 | 13 | The licenses for most software and other practical works are designed 14 | to take away your freedom to share and change the works. By contrast, 15 | the GNU General Public License is intended to guarantee your freedom to 16 | share and change all versions of a program--to make sure it remains free 17 | software for all its users. We, the Free Software Foundation, use the 18 | GNU General Public License for most of our software; it applies also to 19 | any other work released this way by its authors. You can apply it to 20 | your programs, too. 21 | 22 | When we speak of free software, we are referring to freedom, not 23 | price. Our General Public Licenses are designed to make sure that you 24 | have the freedom to distribute copies of free software (and charge for 25 | them if you wish), that you receive source code or can get it if you 26 | want it, that you can change the software or use pieces of it in new 27 | free programs, and that you know you can do these things. 28 | 29 | To protect your rights, we need to prevent others from denying you 30 | these rights or asking you to surrender the rights. Therefore, you have 31 | certain responsibilities if you distribute copies of the software, or if 32 | you modify it: responsibilities to respect the freedom of others. 33 | 34 | For example, if you distribute copies of such a program, whether 35 | gratis or for a fee, you must pass on to the recipients the same 36 | freedoms that you received. You must make sure that they, too, receive 37 | or can get the source code. And you must show them these terms so they 38 | know their rights. 39 | 40 | Developers that use the GNU GPL protect your rights with two steps: 41 | (1) assert copyright on the software, and (2) offer you this License 42 | giving you legal permission to copy, distribute and/or modify it. 43 | 44 | For the developers' and authors' protection, the GPL clearly explains 45 | that there is no warranty for this free software. For both users' and 46 | authors' sake, the GPL requires that modified versions be marked as 47 | changed, so that their problems will not be attributed erroneously to 48 | authors of previous versions. 49 | 50 | Some devices are designed to deny users access to install or run 51 | modified versions of the software inside them, although the manufacturer 52 | can do so. This is fundamentally incompatible with the aim of 53 | protecting users' freedom to change the software. The systematic 54 | pattern of such abuse occurs in the area of products for individuals to 55 | use, which is precisely where it is most unacceptable. Therefore, we 56 | have designed this version of the GPL to prohibit the practice for those 57 | products. If such problems arise substantially in other domains, we 58 | stand ready to extend this provision to those domains in future versions 59 | of the GPL, as needed to protect the freedom of users. 60 | 61 | Finally, every program is threatened constantly by software patents. 62 | States should not allow patents to restrict development and use of 63 | software on general-purpose computers, but in those that do, we wish to 64 | avoid the special danger that patents applied to a free program could 65 | make it effectively proprietary. To prevent this, the GPL assures that 66 | patents cannot be used to render the program non-free. 67 | 68 | The precise terms and conditions for copying, distribution and 69 | modification follow. 70 | 71 | TERMS AND CONDITIONS 72 | 73 | 0. Definitions. 74 | 75 | "This License" refers to version 3 of the GNU General Public License. 76 | 77 | "Copyright" also means copyright-like laws that apply to other kinds of 78 | works, such as semiconductor masks. 79 | 80 | "The Program" refers to any copyrightable work licensed under this 81 | License. Each licensee is addressed as "you". "Licensees" and 82 | "recipients" may be individuals or organizations. 83 | 84 | To "modify" a work means to copy from or adapt all or part of the work 85 | in a fashion requiring copyright permission, other than the making of an 86 | exact copy. The resulting work is called a "modified version" of the 87 | earlier work or a work "based on" the earlier work. 88 | 89 | A "covered work" means either the unmodified Program or a work based 90 | on the Program. 91 | 92 | To "propagate" a work means to do anything with it that, without 93 | permission, would make you directly or secondarily liable for 94 | infringement under applicable copyright law, except executing it on a 95 | computer or modifying a private copy. Propagation includes copying, 96 | distribution (with or without modification), making available to the 97 | public, and in some countries other activities as well. 98 | 99 | To "convey" a work means any kind of propagation that enables other 100 | parties to make or receive copies. Mere interaction with a user through 101 | a computer network, with no transfer of a copy, is not conveying. 102 | 103 | An interactive user interface displays "Appropriate Legal Notices" 104 | to the extent that it includes a convenient and prominently visible 105 | feature that (1) displays an appropriate copyright notice, and (2) 106 | tells the user that there is no warranty for the work (except to the 107 | extent that warranties are provided), that licensees may convey the 108 | work under this License, and how to view a copy of this License. If 109 | the interface presents a list of user commands or options, such as a 110 | menu, a prominent item in the list meets this criterion. 111 | 112 | 1. Source Code. 113 | 114 | The "source code" for a work means the preferred form of the work 115 | for making modifications to it. "Object code" means any non-source 116 | form of a work. 117 | 118 | A "Standard Interface" means an interface that either is an official 119 | standard defined by a recognized standards body, or, in the case of 120 | interfaces specified for a particular programming language, one that 121 | is widely used among developers working in that language. 122 | 123 | The "System Libraries" of an executable work include anything, other 124 | than the work as a whole, that (a) is included in the normal form of 125 | packaging a Major Component, but which is not part of that Major 126 | Component, and (b) serves only to enable use of the work with that 127 | Major Component, or to implement a Standard Interface for which an 128 | implementation is available to the public in source code form. A 129 | "Major Component", in this context, means a major essential component 130 | (kernel, window system, and so on) of the specific operating system 131 | (if any) on which the executable work runs, or a compiler used to 132 | produce the work, or an object code interpreter used to run it. 133 | 134 | The "Corresponding Source" for a work in object code form means all 135 | the source code needed to generate, install, and (for an executable 136 | work) run the object code and to modify the work, including scripts to 137 | control those activities. However, it does not include the work's 138 | System Libraries, or general-purpose tools or generally available free 139 | programs which are used unmodified in performing those activities but 140 | which are not part of the work. For example, Corresponding Source 141 | includes interface definition files associated with source files for 142 | the work, and the source code for shared libraries and dynamically 143 | linked subprograms that the work is specifically designed to require, 144 | such as by intimate data communication or control flow between those 145 | subprograms and other parts of the work. 146 | 147 | The Corresponding Source need not include anything that users 148 | can regenerate automatically from other parts of the Corresponding 149 | Source. 150 | 151 | The Corresponding Source for a work in source code form is that 152 | same work. 153 | 154 | 2. Basic Permissions. 155 | 156 | All rights granted under this License are granted for the term of 157 | copyright on the Program, and are irrevocable provided the stated 158 | conditions are met. This License explicitly affirms your unlimited 159 | permission to run the unmodified Program. The output from running a 160 | covered work is covered by this License only if the output, given its 161 | content, constitutes a covered work. This License acknowledges your 162 | rights of fair use or other equivalent, as provided by copyright law. 163 | 164 | You may make, run and propagate covered works that you do not 165 | convey, without conditions so long as your license otherwise remains 166 | in force. You may convey covered works to others for the sole purpose 167 | of having them make modifications exclusively for you, or provide you 168 | with facilities for running those works, provided that you comply with 169 | the terms of this License in conveying all material for which you do 170 | not control copyright. Those thus making or running the covered works 171 | for you must do so exclusively on your behalf, under your direction 172 | and control, on terms that prohibit them from making any copies of 173 | your copyrighted material outside their relationship with you. 174 | 175 | Conveying under any other circumstances is permitted solely under 176 | the conditions stated below. Sublicensing is not allowed; section 10 177 | makes it unnecessary. 178 | 179 | 3. Protecting Users' Legal Rights From Anti-Circumvention Law. 180 | 181 | No covered work shall be deemed part of an effective technological 182 | measure under any applicable law fulfilling obligations under article 183 | 11 of the WIPO copyright treaty adopted on 20 December 1996, or 184 | similar laws prohibiting or restricting circumvention of such 185 | measures. 186 | 187 | When you convey a covered work, you waive any legal power to forbid 188 | circumvention of technological measures to the extent such circumvention 189 | is effected by exercising rights under this License with respect to 190 | the covered work, and you disclaim any intention to limit operation or 191 | modification of the work as a means of enforcing, against the work's 192 | users, your or third parties' legal rights to forbid circumvention of 193 | technological measures. 194 | 195 | 4. Conveying Verbatim Copies. 196 | 197 | You may convey verbatim copies of the Program's source code as you 198 | receive it, in any medium, provided that you conspicuously and 199 | appropriately publish on each copy an appropriate copyright notice; 200 | keep intact all notices stating that this License and any 201 | non-permissive terms added in accord with section 7 apply to the code; 202 | keep intact all notices of the absence of any warranty; and give all 203 | recipients a copy of this License along with the Program. 204 | 205 | You may charge any price or no price for each copy that you convey, 206 | and you may offer support or warranty protection for a fee. 207 | 208 | 5. Conveying Modified Source Versions. 209 | 210 | You may convey a work based on the Program, or the modifications to 211 | produce it from the Program, in the form of source code under the 212 | terms of section 4, provided that you also meet all of these conditions: 213 | 214 | a) The work must carry prominent notices stating that you modified 215 | it, and giving a relevant date. 216 | 217 | b) The work must carry prominent notices stating that it is 218 | released under this License and any conditions added under section 219 | 7. This requirement modifies the requirement in section 4 to 220 | "keep intact all notices". 221 | 222 | c) You must license the entire work, as a whole, under this 223 | License to anyone who comes into possession of a copy. This 224 | License will therefore apply, along with any applicable section 7 225 | additional terms, to the whole of the work, and all its parts, 226 | regardless of how they are packaged. This License gives no 227 | permission to license the work in any other way, but it does not 228 | invalidate such permission if you have separately received it. 229 | 230 | d) If the work has interactive user interfaces, each must display 231 | Appropriate Legal Notices; however, if the Program has interactive 232 | interfaces that do not display Appropriate Legal Notices, your 233 | work need not make them do so. 234 | 235 | A compilation of a covered work with other separate and independent 236 | works, which are not by their nature extensions of the covered work, 237 | and which are not combined with it such as to form a larger program, 238 | in or on a volume of a storage or distribution medium, is called an 239 | "aggregate" if the compilation and its resulting copyright are not 240 | used to limit the access or legal rights of the compilation's users 241 | beyond what the individual works permit. Inclusion of a covered work 242 | in an aggregate does not cause this License to apply to the other 243 | parts of the aggregate. 244 | 245 | 6. Conveying Non-Source Forms. 246 | 247 | You may convey a covered work in object code form under the terms 248 | of sections 4 and 5, provided that you also convey the 249 | machine-readable Corresponding Source under the terms of this License, 250 | in one of these ways: 251 | 252 | a) Convey the object code in, or embodied in, a physical product 253 | (including a physical distribution medium), accompanied by the 254 | Corresponding Source fixed on a durable physical medium 255 | customarily used for software interchange. 256 | 257 | b) Convey the object code in, or embodied in, a physical product 258 | (including a physical distribution medium), accompanied by a 259 | written offer, valid for at least three years and valid for as 260 | long as you offer spare parts or customer support for that product 261 | model, to give anyone who possesses the object code either (1) a 262 | copy of the Corresponding Source for all the software in the 263 | product that is covered by this License, on a durable physical 264 | medium customarily used for software interchange, for a price no 265 | more than your reasonable cost of physically performing this 266 | conveying of source, or (2) access to copy the 267 | Corresponding Source from a network server at no charge. 268 | 269 | c) Convey individual copies of the object code with a copy of the 270 | written offer to provide the Corresponding Source. This 271 | alternative is allowed only occasionally and noncommercially, and 272 | only if you received the object code with such an offer, in accord 273 | with subsection 6b. 274 | 275 | d) Convey the object code by offering access from a designated 276 | place (gratis or for a charge), and offer equivalent access to the 277 | Corresponding Source in the same way through the same place at no 278 | further charge. You need not require recipients to copy the 279 | Corresponding Source along with the object code. If the place to 280 | copy the object code is a network server, the Corresponding Source 281 | may be on a different server (operated by you or a third party) 282 | that supports equivalent copying facilities, provided you maintain 283 | clear directions next to the object code saying where to find the 284 | Corresponding Source. Regardless of what server hosts the 285 | Corresponding Source, you remain obligated to ensure that it is 286 | available for as long as needed to satisfy these requirements. 287 | 288 | e) Convey the object code using peer-to-peer transmission, provided 289 | you inform other peers where the object code and Corresponding 290 | Source of the work are being offered to the general public at no 291 | charge under subsection 6d. 292 | 293 | A separable portion of the object code, whose source code is excluded 294 | from the Corresponding Source as a System Library, need not be 295 | included in conveying the object code work. 296 | 297 | A "User Product" is either (1) a "consumer product", which means any 298 | tangible personal property which is normally used for personal, family, 299 | or household purposes, or (2) anything designed or sold for incorporation 300 | into a dwelling. In determining whether a product is a consumer product, 301 | doubtful cases shall be resolved in favor of coverage. For a particular 302 | product received by a particular user, "normally used" refers to a 303 | typical or common use of that class of product, regardless of the status 304 | of the particular user or of the way in which the particular user 305 | actually uses, or expects or is expected to use, the product. A product 306 | is a consumer product regardless of whether the product has substantial 307 | commercial, industrial or non-consumer uses, unless such uses represent 308 | the only significant mode of use of the product. 309 | 310 | "Installation Information" for a User Product means any methods, 311 | procedures, authorization keys, or other information required to install 312 | and execute modified versions of a covered work in that User Product from 313 | a modified version of its Corresponding Source. The information must 314 | suffice to ensure that the continued functioning of the modified object 315 | code is in no case prevented or interfered with solely because 316 | modification has been made. 317 | 318 | If you convey an object code work under this section in, or with, or 319 | specifically for use in, a User Product, and the conveying occurs as 320 | part of a transaction in which the right of possession and use of the 321 | User Product is transferred to the recipient in perpetuity or for a 322 | fixed term (regardless of how the transaction is characterized), the 323 | Corresponding Source conveyed under this section must be accompanied 324 | by the Installation Information. But this requirement does not apply 325 | if neither you nor any third party retains the ability to install 326 | modified object code on the User Product (for example, the work has 327 | been installed in ROM). 328 | 329 | The requirement to provide Installation Information does not include a 330 | requirement to continue to provide support service, warranty, or updates 331 | for a work that has been modified or installed by the recipient, or for 332 | the User Product in which it has been modified or installed. Access to a 333 | network may be denied when the modification itself materially and 334 | adversely affects the operation of the network or violates the rules and 335 | protocols for communication across the network. 336 | 337 | Corresponding Source conveyed, and Installation Information provided, 338 | in accord with this section must be in a format that is publicly 339 | documented (and with an implementation available to the public in 340 | source code form), and must require no special password or key for 341 | unpacking, reading or copying. 342 | 343 | 7. Additional Terms. 344 | 345 | "Additional permissions" are terms that supplement the terms of this 346 | License by making exceptions from one or more of its conditions. 347 | Additional permissions that are applicable to the entire Program shall 348 | be treated as though they were included in this License, to the extent 349 | that they are valid under applicable law. If additional permissions 350 | apply only to part of the Program, that part may be used separately 351 | under those permissions, but the entire Program remains governed by 352 | this License without regard to the additional permissions. 353 | 354 | When you convey a copy of a covered work, you may at your option 355 | remove any additional permissions from that copy, or from any part of 356 | it. (Additional permissions may be written to require their own 357 | removal in certain cases when you modify the work.) You may place 358 | additional permissions on material, added by you to a covered work, 359 | for which you have or can give appropriate copyright permission. 360 | 361 | Notwithstanding any other provision of this License, for material you 362 | add to a covered work, you may (if authorized by the copyright holders of 363 | that material) supplement the terms of this License with terms: 364 | 365 | a) Disclaiming warranty or limiting liability differently from the 366 | terms of sections 15 and 16 of this License; or 367 | 368 | b) Requiring preservation of specified reasonable legal notices or 369 | author attributions in that material or in the Appropriate Legal 370 | Notices displayed by works containing it; or 371 | 372 | c) Prohibiting misrepresentation of the origin of that material, or 373 | requiring that modified versions of such material be marked in 374 | reasonable ways as different from the original version; or 375 | 376 | d) Limiting the use for publicity purposes of names of licensors or 377 | authors of the material; or 378 | 379 | e) Declining to grant rights under trademark law for use of some 380 | trade names, trademarks, or service marks; or 381 | 382 | f) Requiring indemnification of licensors and authors of that 383 | material by anyone who conveys the material (or modified versions of 384 | it) with contractual assumptions of liability to the recipient, for 385 | any liability that these contractual assumptions directly impose on 386 | those licensors and authors. 387 | 388 | All other non-permissive additional terms are considered "further 389 | restrictions" within the meaning of section 10. If the Program as you 390 | received it, or any part of it, contains a notice stating that it is 391 | governed by this License along with a term that is a further 392 | restriction, you may remove that term. If a license document contains 393 | a further restriction but permits relicensing or conveying under this 394 | License, you may add to a covered work material governed by the terms 395 | of that license document, provided that the further restriction does 396 | not survive such relicensing or conveying. 397 | 398 | If you add terms to a covered work in accord with this section, you 399 | must place, in the relevant source files, a statement of the 400 | additional terms that apply to those files, or a notice indicating 401 | where to find the applicable terms. 402 | 403 | Additional terms, permissive or non-permissive, may be stated in the 404 | form of a separately written license, or stated as exceptions; 405 | the above requirements apply either way. 406 | 407 | 8. Termination. 408 | 409 | You may not propagate or modify a covered work except as expressly 410 | provided under this License. Any attempt otherwise to propagate or 411 | modify it is void, and will automatically terminate your rights under 412 | this License (including any patent licenses granted under the third 413 | paragraph of section 11). 414 | 415 | However, if you cease all violation of this License, then your 416 | license from a particular copyright holder is reinstated (a) 417 | provisionally, unless and until the copyright holder explicitly and 418 | finally terminates your license, and (b) permanently, if the copyright 419 | holder fails to notify you of the violation by some reasonable means 420 | prior to 60 days after the cessation. 421 | 422 | Moreover, your license from a particular copyright holder is 423 | reinstated permanently if the copyright holder notifies you of the 424 | violation by some reasonable means, this is the first time you have 425 | received notice of violation of this License (for any work) from that 426 | copyright holder, and you cure the violation prior to 30 days after 427 | your receipt of the notice. 428 | 429 | Termination of your rights under this section does not terminate the 430 | licenses of parties who have received copies or rights from you under 431 | this License. If your rights have been terminated and not permanently 432 | reinstated, you do not qualify to receive new licenses for the same 433 | material under section 10. 434 | 435 | 9. Acceptance Not Required for Having Copies. 436 | 437 | You are not required to accept this License in order to receive or 438 | run a copy of the Program. Ancillary propagation of a covered work 439 | occurring solely as a consequence of using peer-to-peer transmission 440 | to receive a copy likewise does not require acceptance. However, 441 | nothing other than this License grants you permission to propagate or 442 | modify any covered work. These actions infringe copyright if you do 443 | not accept this License. Therefore, by modifying or propagating a 444 | covered work, you indicate your acceptance of this License to do so. 445 | 446 | 10. Automatic Licensing of Downstream Recipients. 447 | 448 | Each time you convey a covered work, the recipient automatically 449 | receives a license from the original licensors, to run, modify and 450 | propagate that work, subject to this License. You are not responsible 451 | for enforcing compliance by third parties with this License. 452 | 453 | An "entity transaction" is a transaction transferring control of an 454 | organization, or substantially all assets of one, or subdividing an 455 | organization, or merging organizations. If propagation of a covered 456 | work results from an entity transaction, each party to that 457 | transaction who receives a copy of the work also receives whatever 458 | licenses to the work the party's predecessor in interest had or could 459 | give under the previous paragraph, plus a right to possession of the 460 | Corresponding Source of the work from the predecessor in interest, if 461 | the predecessor has it or can get it with reasonable efforts. 462 | 463 | You may not impose any further restrictions on the exercise of the 464 | rights granted or affirmed under this License. For example, you may 465 | not impose a license fee, royalty, or other charge for exercise of 466 | rights granted under this License, and you may not initiate litigation 467 | (including a cross-claim or counterclaim in a lawsuit) alleging that 468 | any patent claim is infringed by making, using, selling, offering for 469 | sale, or importing the Program or any portion of it. 470 | 471 | 11. Patents. 472 | 473 | A "contributor" is a copyright holder who authorizes use under this 474 | License of the Program or a work on which the Program is based. The 475 | work thus licensed is called the contributor's "contributor version". 476 | 477 | A contributor's "essential patent claims" are all patent claims 478 | owned or controlled by the contributor, whether already acquired or 479 | hereafter acquired, that would be infringed by some manner, permitted 480 | by this License, of making, using, or selling its contributor version, 481 | but do not include claims that would be infringed only as a 482 | consequence of further modification of the contributor version. For 483 | purposes of this definition, "control" includes the right to grant 484 | patent sublicenses in a manner consistent with the requirements of 485 | this License. 486 | 487 | Each contributor grants you a non-exclusive, worldwide, royalty-free 488 | patent license under the contributor's essential patent claims, to 489 | make, use, sell, offer for sale, import and otherwise run, modify and 490 | propagate the contents of its contributor version. 491 | 492 | In the following three paragraphs, a "patent license" is any express 493 | agreement or commitment, however denominated, not to enforce a patent 494 | (such as an express permission to practice a patent or covenant not to 495 | sue for patent infringement). To "grant" such a patent license to a 496 | party means to make such an agreement or commitment not to enforce a 497 | patent against the party. 498 | 499 | If you convey a covered work, knowingly relying on a patent license, 500 | and the Corresponding Source of the work is not available for anyone 501 | to copy, free of charge and under the terms of this License, through a 502 | publicly available network server or other readily accessible means, 503 | then you must either (1) cause the Corresponding Source to be so 504 | available, or (2) arrange to deprive yourself of the benefit of the 505 | patent license for this particular work, or (3) arrange, in a manner 506 | consistent with the requirements of this License, to extend the patent 507 | license to downstream recipients. "Knowingly relying" means you have 508 | actual knowledge that, but for the patent license, your conveying the 509 | covered work in a country, or your recipient's use of the covered work 510 | in a country, would infringe one or more identifiable patents in that 511 | country that you have reason to believe are valid. 512 | 513 | If, pursuant to or in connection with a single transaction or 514 | arrangement, you convey, or propagate by procuring conveyance of, a 515 | covered work, and grant a patent license to some of the parties 516 | receiving the covered work authorizing them to use, propagate, modify 517 | or convey a specific copy of the covered work, then the patent license 518 | you grant is automatically extended to all recipients of the covered 519 | work and works based on it. 520 | 521 | A patent license is "discriminatory" if it does not include within 522 | the scope of its coverage, prohibits the exercise of, or is 523 | conditioned on the non-exercise of one or more of the rights that are 524 | specifically granted under this License. You may not convey a covered 525 | work if you are a party to an arrangement with a third party that is 526 | in the business of distributing software, under which you make payment 527 | to the third party based on the extent of your activity of conveying 528 | the work, and under which the third party grants, to any of the 529 | parties who would receive the covered work from you, a discriminatory 530 | patent license (a) in connection with copies of the covered work 531 | conveyed by you (or copies made from those copies), or (b) primarily 532 | for and in connection with specific products or compilations that 533 | contain the covered work, unless you entered into that arrangement, 534 | or that patent license was granted, prior to 28 March 2007. 535 | 536 | Nothing in this License shall be construed as excluding or limiting 537 | any implied license or other defenses to infringement that may 538 | otherwise be available to you under applicable patent law. 539 | 540 | 12. No Surrender of Others' Freedom. 541 | 542 | If conditions are imposed on you (whether by court order, agreement or 543 | otherwise) that contradict the conditions of this License, they do not 544 | excuse you from the conditions of this License. If you cannot convey a 545 | covered work so as to satisfy simultaneously your obligations under this 546 | License and any other pertinent obligations, then as a consequence you may 547 | not convey it at all. For example, if you agree to terms that obligate you 548 | to collect a royalty for further conveying from those to whom you convey 549 | the Program, the only way you could satisfy both those terms and this 550 | License would be to refrain entirely from conveying the Program. 551 | 552 | 13. Use with the GNU Affero General Public License. 553 | 554 | Notwithstanding any other provision of this License, you have 555 | permission to link or combine any covered work with a work licensed 556 | under version 3 of the GNU Affero General Public License into a single 557 | combined work, and to convey the resulting work. The terms of this 558 | License will continue to apply to the part which is the covered work, 559 | but the special requirements of the GNU Affero General Public License, 560 | section 13, concerning interaction through a network will apply to the 561 | combination as such. 562 | 563 | 14. Revised Versions of this License. 564 | 565 | The Free Software Foundation may publish revised and/or new versions of 566 | the GNU General Public License from time to time. Such new versions will 567 | be similar in spirit to the present version, but may differ in detail to 568 | address new problems or concerns. 569 | 570 | Each version is given a distinguishing version number. If the 571 | Program specifies that a certain numbered version of the GNU General 572 | Public License "or any later version" applies to it, you have the 573 | option of following the terms and conditions either of that numbered 574 | version or of any later version published by the Free Software 575 | Foundation. If the Program does not specify a version number of the 576 | GNU General Public License, you may choose any version ever published 577 | by the Free Software Foundation. 578 | 579 | If the Program specifies that a proxy can decide which future 580 | versions of the GNU General Public License can be used, that proxy's 581 | public statement of acceptance of a version permanently authorizes you 582 | to choose that version for the Program. 583 | 584 | Later license versions may give you additional or different 585 | permissions. However, no additional obligations are imposed on any 586 | author or copyright holder as a result of your choosing to follow a 587 | later version. 588 | 589 | 15. Disclaimer of Warranty. 590 | 591 | THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY 592 | APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT 593 | HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY 594 | OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, 595 | THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 596 | PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM 597 | IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF 598 | ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 599 | 600 | 16. Limitation of Liability. 601 | 602 | IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING 603 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS 604 | THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY 605 | GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE 606 | USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF 607 | DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD 608 | PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), 609 | EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF 610 | SUCH DAMAGES. 611 | 612 | 17. Interpretation of Sections 15 and 16. 613 | 614 | If the disclaimer of warranty and limitation of liability provided 615 | above cannot be given local legal effect according to their terms, 616 | reviewing courts shall apply local law that most closely approximates 617 | an absolute waiver of all civil liability in connection with the 618 | Program, unless a warranty or assumption of liability accompanies a 619 | copy of the Program in return for a fee. 620 | 621 | END OF TERMS AND CONDITIONS 622 | 623 | How to Apply These Terms to Your New Programs 624 | 625 | If you develop a new program, and you want it to be of the greatest 626 | possible use to the public, the best way to achieve this is to make it 627 | free software which everyone can redistribute and change under these terms. 628 | 629 | To do so, attach the following notices to the program. It is safest 630 | to attach them to the start of each source file to most effectively 631 | state the exclusion of warranty; and each file should have at least 632 | the "copyright" line and a pointer to where the full notice is found. 633 | 634 | 635 | Copyright (C) 636 | 637 | This program is free software: you can redistribute it and/or modify 638 | it under the terms of the GNU General Public License as published by 639 | the Free Software Foundation, either version 3 of the License, or 640 | (at your option) any later version. 641 | 642 | This program is distributed in the hope that it will be useful, 643 | but WITHOUT ANY WARRANTY; without even the implied warranty of 644 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 645 | GNU General Public License for more details. 646 | 647 | You should have received a copy of the GNU General Public License 648 | along with this program. If not, see . 649 | 650 | Also add information on how to contact you by electronic and paper mail. 651 | 652 | If the program does terminal interaction, make it output a short 653 | notice like this when it starts in an interactive mode: 654 | 655 | Copyright (C) 656 | This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. 657 | This is free software, and you are welcome to redistribute it 658 | under certain conditions; type `show c' for details. 659 | 660 | The hypothetical commands `show w' and `show c' should show the appropriate 661 | parts of the General Public License. Of course, your program's commands 662 | might be different; for a GUI interface, you would use an "about box". 663 | 664 | You should also get your employer (if you work as a programmer) or school, 665 | if any, to sign a "copyright disclaimer" for the program, if necessary. 666 | For more information on this, and how to apply and follow the GNU GPL, see 667 | . 668 | 669 | The GNU General Public License does not permit incorporating your program 670 | into proprietary programs. If your program is a subroutine library, you 671 | may consider it more useful to permit linking proprietary applications with 672 | the library. If this is what you want to do, use the GNU Lesser General 673 | Public License instead of this License. But first, please read 674 | . 675 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Poisson2Sparse 2 | Official Implementation of Poisson2Sparse - MICCAI 2022 3 | 4 | WARNING REPO IS STILL UNDER DEVELOPMENT AND NEEDS TO BE CLEANEDUP 5 | 6 | 7 | # Docker Instructions 8 | After building and running the container activate the conda environment 9 | `conda activate p2s` before running the inference.py script. -------------------------------------------------------------------------------- /environment.yaml: -------------------------------------------------------------------------------- 1 | name: deep-image-prior 2 | channels: 3 | - pytorch 4 | - conda-forge 5 | - defaults 6 | dependencies: 7 | - cudatoolkit=11.3.1=h9edb442_10 8 | - numpy=1.22.4=py39hc58783e_0 9 | - pillow=9.1.1=py39hae2aec6_1 10 | - pip=22.1.2=pyhd8ed1ab_0 11 | - python=3.9.13=h9a8a25e_0_cpython 12 | - pytorch=1.11.0=py3.9_cuda11.3_cudnn8.2.0_0 13 | - pytorch-mutex=1.0=cuda 14 | - torchvision=0.12.0=py39_cu113 15 | - pip: 16 | - icecream==2.1.2 17 | - imageio==2.19.3 18 | - kornia==0.6.5 19 | - matplotlib==3.5.2 20 | - pywavelets==1.3.0 21 | - pyyaml==6.0 22 | - scikit-image==0.19.2 23 | - scipy==1.8.1 24 | - six==1.16.0 25 | - tifffile==2022.5.4 26 | - tqdm==4.64.0 27 | prefix: /home/cta/anaconda3/envs/deep-image-prior 28 | -------------------------------------------------------------------------------- /sample_images/clean_1000iters.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tacalvin/Poisson2Sparse/f2b18acf755a204f44964cdbd8f1467cbcd8f991/sample_images/clean_1000iters.png -------------------------------------------------------------------------------- /sample_images/noisy.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tacalvin/Poisson2Sparse/f2b18acf755a204f44964cdbd8f1467cbcd8f991/sample_images/noisy.png -------------------------------------------------------------------------------- /src/cscnet.py: -------------------------------------------------------------------------------- 1 | # Credit goes to https://github.com/drorsimon/CSCNet 2 | # Simon, Dror, and Michael Elad. "Rethinking the CSC model for natural images." Advances in Neural Information Processing Systems 32 (2019). 3 | 4 | from collections import namedtuple 5 | 6 | 7 | import torch 8 | import torch.nn as nn 9 | import torch.nn.functional as F 10 | 11 | 12 | ListaParams = namedtuple('ListaParams', ['kernel_size', 'num_filters', 'stride', 'unfoldings', 'channels']) 13 | def calc_pad_sizes(I: torch.Tensor, kernel_size: int, stride: int): 14 | left_pad = stride 15 | right_pad = 0 if (I.shape[3] + left_pad - kernel_size) % stride == 0 else stride - ((I.shape[3] + left_pad - kernel_size) % stride) 16 | top_pad = stride 17 | bot_pad = 0 if (I.shape[2] + top_pad - kernel_size) % stride == 0 else stride - ((I.shape[2] + top_pad - kernel_size) % stride) 18 | right_pad += stride 19 | bot_pad += stride 20 | return left_pad, right_pad, top_pad, bot_pad 21 | 22 | class SoftThreshold(nn.Module): 23 | def __init__(self, size, init_threshold=1e-3): 24 | super(SoftThreshold, self).__init__() 25 | self.threshold = nn.Parameter(init_threshold * torch.ones(1,size,1,1)) 26 | 27 | def forward(self, x): 28 | mask1 = (x > self.threshold).float() 29 | mask2 = (x < -self.threshold).float() 30 | out = mask1.float() * (x - self.threshold) 31 | out += mask2.float() * (x + self.threshold) 32 | return out 33 | 34 | 35 | 36 | 37 | 38 | class ConvLista_T(nn.Module): 39 | def __init__(self, params: ListaParams, A=None, B=None, C=None, threshold=1e-2, norm=False): 40 | super(ConvLista_T, self).__init__() 41 | if A is None: 42 | A = torch.randn(params.num_filters, params.channels, params.kernel_size, params.kernel_size) 43 | l = conv_power_method(A, [128, 128], num_iters=20, stride=params.stride) 44 | # l = conv_power_method(A, [28,28], num_iters=200, stride=params.stride) 45 | A /= torch.sqrt(l) 46 | if B is None: 47 | B = torch.clone(A) 48 | if C is None: 49 | C = torch.clone(A) 50 | self.apply_A = torch.nn.ConvTranspose2d(params.num_filters, params.channels, kernel_size=params.kernel_size, 51 | stride=params.stride, bias=False) 52 | self.apply_B = torch.nn.Conv2d(params.channels, params.num_filters, kernel_size=params.kernel_size, stride=params.stride, bias=False) 53 | self.apply_C = torch.nn.ConvTranspose2d(params.num_filters, params.channels, kernel_size=params.kernel_size, 54 | stride=params.stride, bias=False) 55 | self.apply_A.weight.data = A 56 | self.apply_B.weight.data = B 57 | self.apply_C.weight.data = C 58 | self.soft_threshold = SoftThreshold(params.num_filters, threshold) 59 | self.params = params 60 | self.num_iter = params.unfoldings 61 | # self.norm = norm 62 | # if self.norm: 63 | # self.norm_layer = torch.nn.InstanceNorm2d(params.num_filters) 64 | # self.norm_layer = torch.nn. 65 | 66 | def _split_image(self, I): 67 | if self.params.stride == 1: 68 | return I, torch.ones_like(I) 69 | left_pad, right_pad, top_pad, bot_pad = calc_pad_sizes(I, self.params.kernel_size, self.params.stride) 70 | I_batched_padded = torch.zeros(I.shape[0], self.params.stride ** 2, I.shape[1], top_pad + I.shape[2] + bot_pad, 71 | left_pad + I.shape[3] + right_pad).type_as(I) 72 | valids_batched = torch.zeros_like(I_batched_padded) 73 | for num, (row_shift, col_shift) in enumerate([(i, j) for i in range(self.params.stride) for j in range(self.params.stride)]): 74 | I_padded = F.pad(I, pad=( 75 | left_pad - col_shift, right_pad + col_shift, top_pad - row_shift, bot_pad + row_shift), mode='reflect') 76 | valids = F.pad(torch.ones_like(I), pad=( 77 | left_pad - col_shift, right_pad + col_shift, top_pad - row_shift, bot_pad + row_shift), mode='constant') 78 | I_batched_padded[:, num, :, :, :] = I_padded 79 | valids_batched[:, num, :, :, :] = valids 80 | I_batched_padded = I_batched_padded.reshape(-1, *I_batched_padded.shape[2:]) 81 | valids_batched = valids_batched.reshape(-1, *valids_batched.shape[2:]) 82 | return I_batched_padded, valids_batched 83 | 84 | def disable_warmup(self): 85 | self.num_iter = self.params.unfoldings 86 | 87 | def enable_warmup(self): 88 | self.num_iter = 1 89 | 90 | def forward(self, I): 91 | I_batched_padded, valids_batched = self._split_image(I) 92 | conv_input = self.apply_B(I_batched_padded) #encode 93 | gamma_k = self.soft_threshold(conv_input) 94 | # ic(gamma_k.shape) 95 | for k in range(self.num_iter - 1): 96 | x_k = self.apply_A(gamma_k) # decode 97 | # r_k = self.apply_B(x_k-I_batched_padded) #encode 98 | r_k = self.apply_B(x_k-I_batched_padded) #encode 99 | # if self.norm: 100 | # r_k = self.norm_layer(r_k) 101 | #bug? try adding 102 | gamma_k = self.soft_threshold(gamma_k - r_k) 103 | output_all = self.apply_C(gamma_k) 104 | output_cropped = torch.masked_select(output_all, valids_batched.bool()).reshape(I.shape[0], self.params.stride ** 2, *I.shape[1:]) 105 | # if self.return_all: 106 | # return output_cropped 107 | output = output_cropped.mean(dim=1, keepdim=False) 108 | # output = F.relu(output) 109 | return torch.clamp(output,0.0,1.0) 110 | 111 | 112 | 113 | def conv_power_method(D, image_size, num_iters=100, stride=1): 114 | """ 115 | Finds the maximal eigenvalue of D.T.dot(D) using the iterative power method 116 | :param D: 117 | :param num_needles: 118 | :param image_size: 119 | :param patch_size: 120 | :param num_iters: 121 | :return: 122 | """ 123 | needles_shape = [int(((image_size[0] - D.shape[-2])/stride)+1), int(((image_size[1] - D.shape[-1])/stride)+1)] 124 | x = torch.randn(1, D.shape[0], *needles_shape).type_as(D) 125 | for _ in range(num_iters): 126 | c = torch.norm(x.reshape(-1)) 127 | x = x / c 128 | y = F.conv_transpose2d(x, D, stride=stride) 129 | x = F.conv2d(y, D, stride=stride) 130 | return torch.norm(x.reshape(-1)) 131 | 132 | 133 | 134 | 135 | -------------------------------------------------------------------------------- /src/example.yaml: -------------------------------------------------------------------------------- 1 | dev: false 2 | experiment_cfg: 3 | LAM: 2 4 | cuda: true 5 | dataset: 6 | dataset_path: ./data/PINCAT10 7 | extension: png 8 | gtandraw: true 9 | resize: false 10 | input_type: noise 11 | l1: 1.0e-05 #l1 regularization 12 | lr: 0.0001 13 | num_iter: 5500 14 | optimizer: Adam 15 | poisson_loss: true 16 | experiment_pipeline: ours 17 | model_cfg: 18 | channels: 1 19 | kernel_size: 3 20 | norm: false 21 | num_filters: 512 22 | num_iter: 10 23 | stride: 1 24 | threshold: 0.01 25 | output_dir: ./results/PINCAT10/ 26 | -------------------------------------------------------------------------------- /src/fmd_dataloader.py: -------------------------------------------------------------------------------- 1 | import os 2 | 3 | from PIL import Image 4 | 5 | import torch 6 | import torchvision 7 | from torchvision import transforms 8 | import numpy as np 9 | from torchvision.transforms.transforms import CenterCrop 10 | 11 | from glob import glob 12 | 13 | 14 | class FMDDataset(): 15 | def __init__(self, path): 16 | self.path = path 17 | self.gt_root = os.path.join(path, 'gt/') 18 | self.raw_root = os.path.join(path, 'raw/') 19 | self.transforms = transforms.Compose([ 20 | transforms.ToTensor() 21 | ]) 22 | 23 | self.data = [] 24 | 25 | # 20 elements 26 | gt_elements = glob(self.gt_root+'*') 27 | # print(self.gt_root + "*") 28 | # print(gt_elements) 29 | gt_elements.sort() 30 | # print(gt_elements) 31 | raw_elements = glob(self.raw_root + '*') 32 | raw_elements.sort() 33 | for i in range(len(gt_elements)): 34 | self.data.append( { 35 | 'gt': gt_elements[i], 36 | 'raw': raw_elements[i] 37 | }) 38 | 39 | def __len__(self): 40 | return len(self.data) 41 | 42 | def __getitem__(self, index): 43 | # img_files = glob(self.data[index]['raw']+'*png') 44 | img_files = self.data[index]['raw'] 45 | # print(self.data[index]['raw']+'*png') 46 | # img_files.sort() 47 | img = self.transforms(Image.open(img_files)) 48 | gt = self.transforms(Image.open(self.data[index]['gt'] )) 49 | 50 | # print(img, torch.max(img), torch.min(img)) 51 | # quit() 52 | return img, gt 53 | -------------------------------------------------------------------------------- /src/inference.py: -------------------------------------------------------------------------------- 1 | import argparse 2 | import torch 3 | import torch.nn as nn 4 | import torchvision.io as io 5 | import torchvision.transforms.functional as TF 6 | import torch.nn.functional as F 7 | from tqdm import trange 8 | 9 | import os 10 | from neighbor2neighbor import generate_mask_pair, generate_subimages 11 | from model import build_model 12 | from kornia.metrics import psnr 13 | import yaml 14 | 15 | parser = argparse.ArgumentParser() 16 | parser.add_argument( 17 | "--cfg_path", default="src/example.yaml", help="Model and Hyperparamter Config" 18 | ) 19 | parser.add_argument("--input_path", required=True, help="Path to image to denoise") 20 | parser.add_argument("--output_path", required=True, help="Path to save denoised image") 21 | 22 | 23 | 24 | class Loader(yaml.SafeLoader): 25 | 26 | def __init__(self, stream): 27 | 28 | self._root = os.path.split(stream.name)[0] 29 | 30 | super(Loader, self).__init__(stream) 31 | 32 | def include(self, node): 33 | 34 | filename = self.construct_scalar(node) 35 | 36 | with open(filename, 'r') as f: 37 | return yaml.load(f, Loader) 38 | 39 | 40 | Loader.add_constructor('!include', Loader.include) 41 | 42 | 43 | def main(noisy, config, experiment_cfg): 44 | model = build_model(config) 45 | device = "cuda" if torch.cuda.is_available() else "cpu" 46 | model.to(device) 47 | print( 48 | "Number of params: ", 49 | sum(p.numel() for p in model.parameters() if p.requires_grad), 50 | ) 51 | # optimizer 52 | if experiment_cfg["optimizer"] == "Adam": 53 | LR = experiment_cfg["lr"] 54 | optimizer = torch.optim.AdamW(model.parameters(), lr=LR) 55 | 56 | # psnr_list = [] 57 | # loss_list = [] 58 | # ssims_list = [] 59 | exp_weight = 0.99 60 | 61 | out_avg = None 62 | 63 | noisy_in = noisy 64 | noisy_in = noisy_in.to(device) 65 | 66 | H = None 67 | W = None 68 | # if noisy.shape[1] != noisy.shape[2]: 69 | # H = noisy.shape[2] 70 | # W = noisy.shape[3] 71 | # val_size = (max(H, W) + 31) // 32 * 32 72 | # noisy_in = TF.pad( 73 | # noisy, 74 | # (0, 0, val_size - noisy.shape[3], val_size - noisy.shape[2]), 75 | # padding_mode="reflect", 76 | # ) 77 | 78 | t = trange(experiment_cfg["num_iter"]) 79 | pll = nn.PoissonNLLLoss(log_input=False, full=True) 80 | last_net = None 81 | psrn_noisy_last = 0.0 82 | for i in t: 83 | 84 | mask1, mask2 = generate_mask_pair(noisy_in) 85 | mask1 = mask1.to(device) 86 | mask2 = mask2.to(device) 87 | with torch.no_grad(): 88 | noisy_denoised = model(noisy_in) 89 | noisy_denoised = torch.clamp(noisy_denoised, 0.0, 1.0) 90 | 91 | noisy_in_aug = noisy_in.clone() 92 | # ic(noisy_in_aug.shape, mask1.shape, noisy_in.shape) 93 | noisy_sub1 = generate_subimages(noisy_in_aug, mask1) 94 | noisy_sub2 = generate_subimages(noisy_in_aug, mask2) 95 | 96 | noisy_sub1_denoised = generate_subimages(noisy_denoised, mask1) 97 | noisy_sub2_denoised = generate_subimages(noisy_denoised, mask2) 98 | 99 | noisy_output = model(noisy_sub1) 100 | noisy_output = torch.clamp(noisy_output, 0.0, 1.0) 101 | noisy_target = noisy_sub2 102 | 103 | Lambda = experiment_cfg["LAM"] 104 | diff = noisy_output - noisy_target 105 | exp_diff = noisy_sub1_denoised - noisy_sub2_denoised 106 | 107 | if "l1" in experiment_cfg.keys(): 108 | l1_regularization = 0.0 109 | for param in model.parameters(): 110 | l1_regularization += param.abs().sum() 111 | total_loss = experiment_cfg["l1"] * l1_regularization 112 | # else: 113 | if "poisson_loss" in experiment_cfg.keys(): 114 | loss1 = pll(noisy_output, noisy_target) 115 | loss2 = F.l1_loss(noisy_output, noisy_target) 116 | loss1 += loss2 117 | elif "poisson_loss_only" in experiment_cfg.keys(): 118 | loss1 = pll(noisy_output, noisy_target) 119 | elif "l1_loss" in experiment_cfg.keys(): 120 | loss1 = F.l1_loss(noisy_output, noisy_target) 121 | 122 | elif "mse" in experiment_cfg.keys(): 123 | loss1 = torch.mean(diff ** 2) 124 | else: 125 | loss1 = F.l1_loss(noisy_output, noisy_target) 126 | # orch.mean(diff**2) 127 | loss2 = Lambda * torch.mean((diff - exp_diff) ** 2) 128 | 129 | loss = loss1 + loss2 130 | if "l1" in experiment_cfg.keys(): 131 | loss += total_loss 132 | loss.backward() 133 | 134 | with torch.no_grad(): 135 | out_full = model(noisy_in).detach().cpu() 136 | if H is not None: 137 | out_full = out_full[:, :, :H, :W] 138 | if out_avg is None: 139 | out_avg = out_full.detach().cpu() 140 | else: 141 | out_avg = out_avg * exp_weight + out_full * (1 - exp_weight) 142 | out_avg = out_avg.detach().cpu() 143 | noisy_psnr = psnr(out_full, noisy_in.detach().cpu(), max_val=1.0).item() 144 | 145 | if (i + 1) % 50: 146 | if noisy_psnr - psrn_noisy_last < -4 and last_net is not None: 147 | print("Falling back to previous checkpoint.") 148 | 149 | for new_param, net_param in zip(last_net, model.parameters()): 150 | net_param.data.copy_(new_param.cuda()) 151 | 152 | total_loss = total_loss * 0 153 | optimizer.zero_grad() 154 | torch.cuda.empty_cache() 155 | continue 156 | else: 157 | last_net = [x.detach().cpu() for x in model.parameters()] 158 | psrn_noisy_last = noisy_psnr 159 | 160 | optimizer.step() 161 | optimizer.zero_grad() 162 | 163 | with torch.no_grad(): 164 | out_full = model(noisy_in).detach().cpu() 165 | if H is not None: 166 | out_full = out_full[:, :, :H, :W] 167 | if out_avg is None: 168 | out_avg = out_full.detach().cpu() 169 | else: 170 | out_avg = out_avg * exp_weight + out_full * (1 - exp_weight) 171 | out_avg = out_avg.detach().cpu() 172 | 173 | return out_avg 174 | 175 | 176 | if __name__ == "__main__": 177 | args = parser.parse_args() 178 | 179 | with open(args.cfg_path, "r") as f: 180 | cfg = yaml.load(f, Loader=Loader) 181 | 182 | noisy = io.read_image(args.input_path).unsqueeze(0)/255 183 | 184 | out_image = main(noisy, cfg, cfg['experiment_cfg']) * 255 185 | out_image = out_image.type(torch.uint8).squeeze(0) 186 | io.write_png(out_image, args.output_path) -------------------------------------------------------------------------------- /src/model.py: -------------------------------------------------------------------------------- 1 | from cscnet import ConvLista_T, ListaParams 2 | 3 | def build_model(cfg): 4 | params = ListaParams(cfg['model_cfg']['kernel_size'], cfg['model_cfg']['num_filters'], cfg['model_cfg']['stride'], 5 | cfg['model_cfg']['num_iter'], cfg['model_cfg']['channels']) 6 | net = ConvLista_T(params, threshold=cfg['model_cfg']['threshold'], norm=cfg['model_cfg']['norm']) 7 | return net 8 | -------------------------------------------------------------------------------- /src/neighbor2neighbor.py: -------------------------------------------------------------------------------- 1 | # Credit goes to https://github.com/TaoHuang2018/Neighbor2Neighbor 2 | # Huang, Tao, et al. "Neighbor2neighbor: Self-supervised denoising from single noisy images." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021. 3 | from operator import sub 4 | import torch 5 | import numpy as np 6 | 7 | operation_seed_counter = 0 8 | def get_generator(): 9 | global operation_seed_counter 10 | operation_seed_counter += 1 11 | g_cuda_generator = torch.Generator(device="cuda") 12 | g_cuda_generator.manual_seed(operation_seed_counter) 13 | return g_cuda_generator 14 | 15 | class AugmentNoise(object): 16 | def __init__(self, style): 17 | if style.startswith('gauss'): 18 | self.params = [float(p) / 255.0 for p in style.replace('gauss', '', 1).split('_')] 19 | if len(self.params) == 1: 20 | self.style = "gauss_fix" 21 | elif len(self.params) == 2: 22 | self.style = "gauss_range" 23 | elif style.startswith('poisson'): 24 | self.params = [float(p) for p in style.replace('poisson', '', 1).split('_')] 25 | if len(self.params) == 1: 26 | self.style = "poisson_fix" 27 | elif len(self.params) == 2: 28 | self.style = "poisson_range" 29 | print(self.params) 30 | elif style.startswith('gain'): 31 | self.params = [float(p) for p in style.replace('gain', '', 1).split('_')] 32 | if len(self.params) == 1: 33 | self.style = "gain_fix" 34 | elif len(self.params) == 2: 35 | self.style = "gain_range" 36 | def add_train_noise(self, x): 37 | shape = x.shape 38 | if self.style == "gauss_fix": 39 | std = self.params[0] 40 | std = std * torch.ones((shape[0], 1, 1, 1), device=x.device) 41 | noise = torch.cuda.FloatTensor(shape, device=x.device) 42 | torch.normal(mean=0.0, std=std, generator=get_generator(), out=noise) 43 | return x + noise 44 | elif self.style == "gauss_range": 45 | min_std, max_std = self.params 46 | std = torch.rand(size=(shape[0], 1, 1, 1), device=x.device) * (max_std - min_std) + min_std 47 | noise = torch.cuda.FloatTensor(shape, device=x.device) 48 | torch.normal(mean=0, std=std, generator=get_generator(), out=noise) 49 | return x + noise 50 | elif self.style == "poisson_fix": 51 | lam = self.params[0] 52 | lam = lam * torch.ones((shape[0], 1, 1, 1), device=x.device) 53 | noised = torch.poisson(lam * x, generator=get_generator()) / lam 54 | noised = torch.clamp(noised, 0.0,1.0) 55 | return noised 56 | elif self.style == "poisson_range": 57 | min_lam, max_lam = self.params 58 | lam = torch.rand(size=(shape[0], 1, 1, 1), device=x.device) * (max_lam - min_lam) + min_lam 59 | noised = torch.poisson(lam * x, generator=get_generator()) / lam 60 | return noised 61 | elif self.style == "gain_fix": 62 | lam = self.params[0] 63 | lam = lam * torch.ones((shape[0], 1, 1, 1), device=x.device) 64 | noised = torch.poisson(x / lam, generator=get_generator()) * lam 65 | return noised 66 | elif self.style == "gain_range": 67 | min_lam, max_lam = self.params 68 | lam = torch.rand(size=(shape[0], 1, 1, 1), device=x.device) * (max_lam - min_lam) + min_lam 69 | noised = torch.poisson(x/lam, generator=get_generator()) * lam 70 | return noised 71 | def add_valid_noise(self, x): 72 | shape = x.shape 73 | if self.style == "gauss_fix": 74 | std = self.params[0] 75 | return np.array(x + np.random.normal(size=shape) * std, dtype=np.float32) 76 | elif self.style == "gauss_range": 77 | min_std, max_std = self.params 78 | std = np.random.uniform(low=min_std, high=max_std, size=(1, 1, 1)) 79 | return np.array(x + np.random.normal(size=shape) * std, dtype=np.float32) 80 | elif self.style == "poisson_fix": 81 | lam = self.params[0] 82 | return np.array(np.random.poisson(lam * x) / lam, dtype=np.float32) 83 | elif self.style == "poisson_range": 84 | min_lam, max_lam = self.params 85 | lam = np.random.uniform(low=min_lam, high=max_lam, size=(1, 1, 1)) 86 | return np.array(np.random.poisson(lam * x) / lam, dtype=np.float32) 87 | 88 | 89 | def space_to_depth(x, block_size): 90 | n, c, h, w = x.size() 91 | unfolded_x = torch.nn.functional.unfold(x, block_size, stride=block_size) 92 | return unfolded_x.view(n, c * block_size**2, h // block_size, 93 | w // block_size) 94 | 95 | def generate_mask_pair(img): 96 | # prepare masks (N x C x H/2 x W/2) 97 | n, c, h, w = img.shape 98 | mask1 = torch.zeros(size=(n * h // 2 * w // 2 * 4, ), 99 | dtype=torch.bool, 100 | device=img.device) 101 | mask2 = torch.zeros(size=(n * h // 2 * w // 2 * 4, ), 102 | dtype=torch.bool, 103 | device=img.device) 104 | # prepare random mask pairs 105 | idx_pair = torch.tensor( 106 | [[0, 1], [0, 2], [1, 3], [2, 3], [1, 0], [2, 0], [3, 1], [3, 2]], 107 | dtype=torch.int64, 108 | device=img.device) 109 | rd_idx = torch.zeros(size=(n * h // 2 * w // 2, ), 110 | dtype=torch.int64, 111 | device=img.device) 112 | torch.randint(low=0, 113 | high=8, 114 | size=(n * h // 2 * w // 2, ), 115 | generator=get_generator(), 116 | out=rd_idx) 117 | rd_pair_idx = idx_pair[rd_idx] 118 | rd_pair_idx += torch.arange(start=0, 119 | end=n * h // 2 * w // 2 * 4, 120 | step=4, 121 | dtype=torch.int64, 122 | device=img.device).reshape(-1, 1) 123 | # get masks 124 | mask1[rd_pair_idx[:, 0]] = 1 125 | mask2[rd_pair_idx[:, 1]] = 1 126 | return mask1, mask2 127 | 128 | def generate_subimages(img, mask): 129 | n, c, h, w = img.shape 130 | subimage = torch.zeros(n, 131 | c, 132 | h // 2, 133 | w // 2, 134 | dtype=img.dtype, 135 | layout=img.layout, 136 | device=img.device) 137 | # per channel 138 | for i in range(c): 139 | img_per_channel = space_to_depth(img[:, i:i + 1, :, :], block_size=2) 140 | # ic(img_per_channel.shape, subimage.shape) 141 | img_per_channel = img_per_channel.permute(0, 2, 3, 1).reshape(-1) 142 | channel_mask = img_per_channel[mask].reshape( 143 | n, h // 2, w // 2, 1).permute(0, 3, 1, 2) 144 | # ic(channel_mask.shape, subimage.shape) 145 | subimage[:, i:i+1, :, :] = channel_mask 146 | return subimage -------------------------------------------------------------------------------- /src/run_folder.py: -------------------------------------------------------------------------------- 1 | import argparse 2 | import os 3 | import pickle 4 | import glob 5 | import random 6 | 7 | import kornia 8 | from kornia.losses import psnr 9 | 10 | import numpy as np 11 | from numpy.core.fromnumeric import resize, transpose 12 | import torch.nn.functional as F 13 | from torch.utils.data import DataLoader 14 | import torchvision.datasets as datasets 15 | from torchvision.transforms import transforms 16 | import torchvision.transforms.functional as T 17 | from training import train 18 | from fmd_dataloader import FMDDataset 19 | 20 | # from tqdm import tqdm 21 | from tqdm import trange 22 | 23 | import torch 24 | import torch.optim 25 | 26 | from skimage.metrics import peak_signal_noise_ratio 27 | from skimage.util import random_noise 28 | import yaml 29 | import datetime 30 | 31 | from PIL import Image 32 | 33 | # from utils.denoising_utils import * 34 | 35 | torch.manual_seed(123) 36 | np.random.seed(123) 37 | 38 | 39 | 40 | 41 | def load_dataset(cfg_experiment): 42 | # print(cfg_experiment) 43 | path = cfg_experiment['dataset']['dataset_path'] 44 | # print(path) 45 | 46 | # if we have gt and raw we need to only load gt 47 | if 'gtandraw' in cfg_experiment['dataset'].keys() and cfg_experiment['dataset']['gtandraw']: 48 | print("path", path) 49 | return DataLoader(FMDDataset(path), batch_size=1, shuffle=False) 50 | 51 | transform_list = [] 52 | if cfg_experiment['dataset']['resize']: 53 | 54 | transform_list.append(transforms.Resize((256,256))) 55 | if cfg_experiment['dataset']['greyscale']: 56 | transform_list.append(transforms.Grayscale()) 57 | 58 | transform_list.append(transforms.ToTensor()) 59 | 60 | transform = transforms.Compose(transform_list) 61 | if 'mnist' in cfg_experiment['dataset'].keys(): 62 | dataset = datasets.MNIST("~", train=False, transform=transform) 63 | dataset = torch.utils.data.Subset(dataset, np.random.choice(np.arange(len(dataset)), size=32)) 64 | imgs = DataLoader(dataset, batch_size=1, shuffle=True) 65 | else: 66 | dataset = datasets.ImageFolder(os.path.dirname(path),transform) 67 | imgs = DataLoader(dataset, batch_size=1, shuffle=False) 68 | 69 | return imgs 70 | 71 | 72 | # 73 | 74 | class Loader(yaml.SafeLoader): 75 | 76 | def __init__(self, stream): 77 | 78 | self._root = os.path.split(stream.name)[0] 79 | 80 | super(Loader, self).__init__(stream) 81 | 82 | def include(self, node): 83 | 84 | filename = self.construct_scalar(node) 85 | 86 | with open(filename, 'r') as f: 87 | return yaml.load(f, Loader) 88 | 89 | 90 | Loader.add_constructor('!include', Loader.include) 91 | 92 | # def p_noise(img, PEAK): 93 | 94 | # return random_noise(img, 'poisson') 95 | 96 | # def p_noise(img, PEAK): 97 | # # print(img, PEAK) 98 | # Q = np.max(np.max(img)) / PEAK 99 | # rate = img / Q 100 | # noisy = np.random.poisson(rate) * Q 101 | 102 | # # print(noisy) 103 | # # quit() 104 | # return noisy 105 | 106 | 107 | def p_noise(img, PEAK): 108 | img = np.multiply(img, PEAK) 109 | img = np.random.poisson(img) 110 | img = np.divide(img, PEAK) 111 | return img 112 | 113 | 114 | def g_noise(img, sigma): 115 | return random_noise(img, 'gaussian', var=float((sigma/255.0)**2)) 116 | 117 | 118 | def gp_noise(img, PEAK, sigma): 119 | return random_noise(p_noise(img, PEAK), 'gaussian', var=sigma/255) 120 | 121 | 122 | def corrupt_dataset(imgs, cfg): 123 | if cfg['noise'] == 'p': 124 | # print() 125 | return [np.clip(p_noise(img, cfg['peak']), 0,1.0) for img in imgs] 126 | elif cfg['noise'] == 'g': 127 | return [np.clip(g_noise(img, cfg['sigma']), 0, 1.0) for img in imgs] 128 | elif cfg['noise'] == 'gp': 129 | return [np.clip(gp_noise(img, cfg['peak'], cfg['sigma']),0,1.0) for img in imgs] 130 | else: 131 | return imgs 132 | 133 | # return imgs 134 | 135 | 136 | def denoise_experiment(cfg): 137 | # load images 138 | output_path = cfg['output_dir'] 139 | experiment_cfg = cfg['experiment_cfg'] 140 | imgs = load_dataset(experiment_cfg) 141 | 142 | # if experiment_cfg['pipeline'] == 'nb': 143 | train(imgs, cfg) 144 | 145 | # print(noisy_imgs[0]) 146 | 147 | 148 | def create_result_dir(cfg, dev=True): 149 | print(cfg) 150 | path = cfg['output_dir'] 151 | try: 152 | num_exp = os.listdir(path) 153 | except: 154 | num_exp = [] 155 | # print(num_exp) 156 | curr_dir_id = len(num_exp) 157 | output_path = os.path.join(path, "{:04d}".format(curr_dir_id+1)) 158 | print(output_path) 159 | try: 160 | # os.mkdir(output_path) 161 | os.makedirs(output_path) 162 | except: 163 | pass 164 | 165 | # create copy of cfg into dir 166 | 167 | with open(os.path.join(output_path, 'cfg.yaml'), 'w') as yaml_file: 168 | yaml.dump(cfg, yaml_file, default_flow_style=False) 169 | 170 | return output_path 171 | 172 | 173 | def experiment(cfg): 174 | start_time = datetime.datetime.now() 175 | print("Begining Experiment {}".format(start_time)) 176 | output_dir = create_result_dir(cfg, dev=cfg['dev']) 177 | if output_dir is not None: 178 | cfg['output_dir'] = output_dir 179 | 180 | # run experiment here 181 | denoise_experiment(cfg) 182 | 183 | end_time = datetime.datetime.now() 184 | print("Ending Experiment {}".format(end_time)) 185 | 186 | 187 | def main(cfg_path): 188 | # load path 189 | # import yaml 190 | with open(cfg_path, 'r') as f: 191 | cfg = yaml.load(f, Loader=Loader) 192 | experiment(cfg) 193 | 194 | 195 | if __name__ == "__main__": 196 | parser = argparse.ArgumentParser(description='Run Denoising Experiment') 197 | parser.add_argument('--cfg_path', help='Path to experiment config') 198 | 199 | args = parser.parse_args() 200 | 201 | main(args.cfg_path) 202 | -------------------------------------------------------------------------------- /src/training.py: -------------------------------------------------------------------------------- 1 | import os 2 | import pickle 3 | import glob 4 | import random 5 | 6 | import kornia 7 | import torch 8 | import torch.nn as nn 9 | import torch.nn.functional as F 10 | # from kornia.losses import psnr_loss as psnr 11 | from kornia.metrics import psnr 12 | import torchvision 13 | 14 | from torchvision.transforms import functional as TF 15 | from model import build_model 16 | from icecream import ic 17 | 18 | 19 | from tqdm import trange 20 | 21 | 22 | from neighbor2neighbor import AugmentNoise, generate_mask_pair, generate_subimages 23 | # from utils.denoising_utils import * 24 | 25 | def build_loss(cfg): 26 | if cfg['experiment_cfg']['mse']: 27 | loss = torch.nn.MSELoss() 28 | elif cfg['experiment_cfg']['pll']: 29 | loss = torch.nn.PoissonNLLLoss() 30 | return loss 31 | 32 | def train(dloader, config): 33 | # torch.cuda.set_per_process_memory_fraction(.6) 34 | experiment_cfg = config['experiment_cfg'] 35 | output_path = config['output_dir'] 36 | 37 | if config['experiment_cfg']['cuda']: 38 | dtype = torch.cuda.FloatTensor 39 | else: 40 | dtype = torch.FloatTensor 41 | 42 | # model = build_model(config) 43 | # .type(dtype) 44 | 45 | # init_state_dict = model.state_dict() 46 | # model.type(dtype) 47 | # create noise generator 48 | # noise_adder = AugmentNoise(style=experiment_cfg['noise']) 49 | running_psnr_avg = 0.0 50 | running_ssim_avg = 0.0 51 | # print(config) 52 | print(len(dloader)) 53 | for idx, img in enumerate(dloader): 54 | if 'gtandraw' in experiment_cfg['dataset'].keys() and experiment_cfg['dataset']['gtandraw']: 55 | noisy, img = img 56 | print(type(img), len(img)) 57 | img = img.type(dtype) 58 | noisy = noisy.type(dtype) 59 | else: 60 | noise_adder = AugmentNoise(style=experiment_cfg['noise']) 61 | img, _ = img 62 | img = img.type(dtype) 63 | 64 | 65 | #noisy image 66 | noisy = noise_adder.add_train_noise(img).type(dtype) 67 | 68 | # with profile(activities=[ProfilerActivity.CUDA],profile_memory=True, record_shapes=True) as prof: 69 | # with record_function("model_inference"): 70 | 71 | 72 | results = train_helper(img, noisy, dtype, config, experiment_cfg) 73 | # ic(results) 74 | denoised, clean_psnr, psnr_list, loss_list, lpips_list, ssims_list = results 75 | # print(prof.key_averages().table(sort_by="cuda_time_total", row_limit=10)) 76 | # print(prof.key_averages().table(sort_by="self_gpu_memory_usage", row_limit=10)) 77 | # print(prof.key_averages().table(sort_by="gpu_memory_usage", row_limit=10)) 78 | image_results_path = os.path.join(output_path, str(idx)) 79 | try: 80 | print(image_results_path) 81 | os.mkdir(image_results_path) 82 | except: 83 | pass 84 | # write clean and noisy image 85 | torchvision.utils.save_image(img, os.path.join(image_results_path,"{}_gt.png".format(idx))) 86 | torchvision.utils.save_image(denoised, os.path.join(image_results_path,"{}_{:.3f}_out.png".format(idx, clean_psnr))) 87 | torchvision.utils.save_image(noisy, os.path.join(image_results_path,"{}_noisy.png".format(idx))) 88 | 89 | # torch.save(model) 90 | with open(os.path.join(image_results_path, 'metrics_psnr.pkl'), 'wb') as file: 91 | # yaml.dump(result_i['metrics'], yaml_file, default_flow_style=False) 92 | pickle.dump({'loss':loss_list, 'psnr':psnr_list, 'lpips':lpips_list, 'ssims': ssims_list}, file) 93 | running_psnr_avg += clean_psnr 94 | 95 | torch.cuda.empty_cache() 96 | #clean up 97 | # del model 98 | # model = build_model(config) 99 | # model.load_state_dict(init_state_dict) 100 | # model.type(dtype) 101 | 102 | print("#############################\n Final Average PSNR: {} SSIM:{}".format(running_psnr_avg/ len(dloader), running_ssim_avg/ len(dloader))) 103 | 104 | 105 | 106 | 107 | 108 | def train_helper( img, noisy, dtype, config, experiment_cfg): 109 | 110 | return nb2nb_aug_helper( img, noisy, dtype, config, experiment_cfg) 111 | 112 | def nb2nb_aug_helper( img, noisy, dtype, config, experiment_cfg): 113 | model = build_model(config) 114 | model.type(dtype) 115 | 116 | print("Number of params: ", sum(p.numel() for p in model.parameters() if p.requires_grad)) 117 | # swa_model = AveragedModel(model) 118 | # loss 119 | # loss_fn_alex = lpips.LPIPS(net='alex') 120 | # mse = torch.nn.MSELoss().type(dtype) 121 | # optimizer 122 | if experiment_cfg['optimizer'] == 'Adam': 123 | LR = experiment_cfg['lr'] 124 | optimizer = torch.optim.AdamW( 125 | model.parameters(), lr=LR) 126 | 127 | if 'lr_sched' in experiment_cfg.keys(): 128 | scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=100, gamma=.99) 129 | # swa_scheduler = SWALR(optimizer, swa_lr=0.05) 130 | 131 | psnr_list = [] 132 | loss_list = [] 133 | lpips_list =[] 134 | ssims_list = [] 135 | grad_hist = [] 136 | exp_weight = .99 137 | 138 | out_avg = None 139 | # optimize single image 140 | 141 | 142 | 143 | 144 | noisy_in = noisy 145 | # WITH CSCNET did nothing somehow 146 | if 'rotate' in experiment_cfg.keys(): 147 | noisy_in = torch.cat((noisy, TF.rotate(noisy, 90), TF.rotate(noisy, 180), TF.rotate(noisy, 270))) 148 | 149 | if 'flip' in experiment_cfg.keys(): 150 | noisy_in = torch.cat((noisy_in, TF.hflip(noisy_in), TF.vflip(noisy_in))) 151 | #horizonal and vertical flip 152 | #we need to pad to a square if the image is not already a square 153 | H = None 154 | W = None 155 | if noisy.shape[2] != noisy.shape[3]: 156 | H = noisy.shape[2] 157 | W = noisy.shape[3] 158 | val_size = (max(H, W) + 31) // 32 * 32 159 | noisy_in = TF.pad(noisy, (0, 0, val_size-noisy.shape[3], val_size-noisy.shape[2]), padding_mode='reflect') 160 | # noisy_preaug = noisy_in 161 | t = trange(experiment_cfg['num_iter']) 162 | pll = nn.PoissonNLLLoss(log_input=False, full=True) 163 | last_net = None 164 | grad_hist = [] 165 | # model.enable_warmup() 166 | # dict_checkpointC = model.apply_C.weight.data.clone().cpu() 167 | # dict_checkpointA = model.apply_A.weight.data.clone().cpu() 168 | # dict_checkpointB = model.apply_B.weight.data.clone().cpu() 169 | warmup = True 170 | warmup_counter = 0 171 | psrn_noisy_last =0.0 172 | for i in t: 173 | 174 | # if i==1000 or (i > 1000 and warmup and warmup_counter == 50 ): 175 | # dict_checkpointC = model.apply_C.weight.data.clone().cpu() 176 | # dict_checkpointA = model.apply_A.weight.data.clone().cpu() 177 | # dict_checkpointB = model.apply_B.weight.data.clone().cpu() 178 | # warmup_counter = 0 179 | # warmup = False 180 | # # optimizer.param_groups[0]['lr'] = optimizer.param_groups[0]['lr']*.1 181 | # model.disable_warmup() 182 | # print("Disable Warmup: {}".format(i)) 183 | # elif i > 1000 and warmup_counter == 200 and not warmup: 184 | # warmup_counter = 0 185 | # warmup = True 186 | # model.enable_warmup() 187 | # print("Enable Warmup: {}".format(i)) 188 | # warmup_counter += 1 189 | mask1, mask2 = generate_mask_pair(noisy_in) 190 | #g1(y) #g2(y) 191 | # noisy_sub1 = generate_subimages(noisy_in, mask1) 192 | # noisy_sub2 = generate_subimages(noisy_in, mask2) 193 | 194 | 195 | 196 | # if experiment_cfg['regularizer']: 197 | with torch.no_grad(): 198 | # if out_avg is None: 199 | # if config['model_cfg']['model_type'] == 'deepcdl': 200 | # noisy_denoised = model(noisy_in, torch.Tensor([50/255]).unsqueeze(1).unsqueeze(1).unsqueeze(1).type(dtype)) 201 | # else: 202 | noisy_denoised = model(noisy_in) 203 | noisy_denoised = torch.clamp(noisy_denoised, 0.0, 1.0) 204 | 205 | # else: 206 | # noisy_denoised - out_avg.clone().type(dtype) 207 | 208 | if 'cutnoise' in experiment_cfg.keys(): 209 | noisy_denoised, noisy_in_aug = cutNoise(noisy_denoised.clone(), noisy_in.clone()) 210 | else: 211 | noisy_in_aug = noisy_in.clone() 212 | # ic(noisy_in_aug.shape, mask1.shape, noisy_in.shape) 213 | noisy_sub1 = generate_subimages(noisy_in_aug, mask1) 214 | noisy_sub2 = generate_subimages(noisy_in_aug, mask2) 215 | 216 | 217 | #TODO Add noise to sub1? 218 | 219 | # ic(noisy_denoised.shape) 220 | noisy_sub1_denoised = generate_subimages(noisy_denoised, mask1) 221 | noisy_sub2_denoised = generate_subimages(noisy_denoised, mask2) 222 | # ic(config) 223 | # if config['model_cfg']['model_type'] == 'deepcdl': 224 | # noisy_output = model(noisy_sub1, torch.Tensor([50/255]).unsqueeze(1).unsqueeze(1).unsqueeze(1).type(dtype)) 225 | # else: 226 | noisy_output = model(noisy_sub1) 227 | # print("MODEL: {}".format(i)) 228 | noisy_output = torch.clamp(noisy_output, 0.0, 1.0) 229 | # if H is not None: 230 | # noisy_output = noisy_output[:,:, :H, :W] 231 | noisy_target = noisy_sub2 232 | 233 | # ic(noisy_output.shape) 234 | Lambda = experiment_cfg['LAM'] 235 | # Lambda = i /experiment_cfg['num_iter'] * experiment_cfg['LAM'] 236 | # ic(noisy_output.shape, noisy_target.shape) 237 | diff = noisy_output - noisy_target 238 | exp_diff = noisy_sub1_denoised - noisy_sub2_denoised 239 | # if cfg['experiment_cfg']['loss'] == 'poisson': 240 | # total_loss = pll(out, img_noisy_torch) 241 | if "l1" in experiment_cfg.keys(): 242 | l1_regularization = 0. 243 | for param in model.parameters(): 244 | l1_regularization += param.abs().sum() 245 | total_loss = (experiment_cfg['l1'] * l1_regularization) 246 | # else: 247 | if 'poisson_loss' in experiment_cfg.keys(): 248 | loss1 = pll(noisy_output, noisy_target) 249 | loss2 = F.l1_loss(noisy_output, noisy_target) 250 | loss1 += loss2 251 | # gamma = .5 252 | # loss1 = (gamma *loss1) + ((1-gamma) * loss2) 253 | elif 'poisson_loss_only' in experiment_cfg.keys(): 254 | loss1 = pll(noisy_output, noisy_target) 255 | # loss2 = F.l1_loss(noisy_output, noisy_target) 256 | # loss1 += loss2 257 | # gamma = .5 258 | # loss1 = (gamma *loss1) + ((1-gamma) * loss2) 259 | elif 'l1_loss' in experiment_cfg.keys(): 260 | # loss1 = pll(noisy_output, noisy_target) 261 | loss1 = F.l1_loss(noisy_output, noisy_target) 262 | # loss1 += loss2 263 | elif 'sure_loss' in experiment_cfg.keys(): 264 | n = torch.randn((noisy_output.shape), requires_grad=True).type(dtype) 265 | div = (n@noisy_output).sum() 266 | div = torch.autograd.grad(div, noisy_output, retain_graph=True)[0] 267 | 268 | loss1 = F.l1_loss(noisy_output, noisy_target) + (n @ div).mean() 269 | 270 | elif 'wsure' in experiment_cfg.keys(): 271 | # fidelity_loss = F.l1_loss(noisy_output, noisy_target) 272 | fidelity_loss = torch.mean(diff**2) 273 | epsilon = 1e-3 274 | eta = noisy_sub1.clone().normal_() 275 | net_input_perturbed = noisy_sub1.clone() + (eta * epsilon) 276 | out_perturbed = model (net_input_perturbed) 277 | dx = out_perturbed - noisy_output 278 | eta_dx = torch.sum(eta * dx) 279 | MCdiv = eta_dx / epsilon 280 | div_term = 2. * (50/255) ** 2 * MCdiv / torch.numel(noisy_sub1) 281 | loss1 = fidelity_loss - (50/255) **2 + div_term 282 | elif "mse" in experiment_cfg.keys(): 283 | loss1 = torch.mean(diff**2) 284 | else: 285 | loss1 = F.l1_loss(noisy_output, noisy_target) 286 | # orch.mean(diff**2) 287 | # loss1 = F.poisson_nll_loss(noisy_output, noisy_target, log_input=False) 288 | # loss1 = torch.nn.functional.l1_loss(noisy_output, noisy_target) 289 | loss2 = Lambda * torch.mean((diff - exp_diff)**2) 290 | 291 | loss = loss1 + loss2 292 | if "l1" in experiment_cfg.keys(): 293 | loss += total_loss 294 | loss.backward() 295 | 296 | 297 | 298 | with torch.no_grad(): 299 | # if i > 500: 300 | # out_full = swa_model(noisy).detach().cpu() 301 | # else: 302 | out_full = model(noisy).detach().cpu() 303 | if H is not None: 304 | out_full = out_full[:,:, :H, :W] 305 | if out_avg is None: 306 | out_avg = out_full.detach().cpu() 307 | else: 308 | out_avg = out_avg * exp_weight + out_full * (1 - exp_weight) 309 | out_avg = out_avg.detach().cpu() 310 | clean_psnr = psnr(out_full, img.detach().cpu(), max_val=1.0).item() 311 | noisy_psnr = psnr(out_full, noisy.detach().cpu(), max_val=1.0).item() 312 | clean_psnr_avg = psnr(out_avg, img.detach().cpu(), max_val=1.0).item() 313 | 314 | 315 | if (i+1) % 50: 316 | if noisy_psnr - psrn_noisy_last < -4 and last_net is not None: 317 | print('Falling back to previous checkpoint.') 318 | 319 | for new_param, net_param in zip(last_net, model.parameters()): 320 | net_param.data.copy_(new_param.cuda()) 321 | 322 | total_loss = total_loss*0 323 | optimizer.zero_grad() 324 | torch.cuda.empty_cache() 325 | continue 326 | else: 327 | last_net = [x.detach().cpu() for x in model.parameters()] 328 | psrn_noisy_last = noisy_psnr 329 | 330 | 331 | 332 | 333 | optimizer.step() 334 | if 'param_noise_sigma' in experiment_cfg.keys(): 335 | add_noise(model, experiment_cfg 336 | ['param_noise_sigma'], learning_rate=LR, dtype=dtype) 337 | # new_model_params = [p.grad.data.clone().detach().cpu() for p in model.parameters()] 338 | # if i > 500: 339 | # swa_model.update_parameters(model) 340 | # swa_scheduler.step() 341 | optimizer.zero_grad() 342 | 343 | if 'lr_sched' in experiment_cfg.keys(): 344 | scheduler.step() 345 | 346 | 347 | with torch.no_grad(): 348 | # if i > 500: 349 | # out_full = swa_model(noisy).detach().cpu() 350 | # else: 351 | out_full = model(noisy).detach().cpu() 352 | if H is not None: 353 | out_full = out_full[:,:, :H, :W] 354 | if out_avg is None: 355 | out_avg = out_full.detach().cpu() 356 | else: 357 | out_avg = out_avg * exp_weight + out_full * (1 - exp_weight) 358 | out_avg = out_avg.detach().cpu() 359 | clean_psnr = psnr(out_full, img.detach().cpu(), max_val=1.0).item() 360 | clean_psnr_avg = psnr(out_avg, img.detach().cpu(), max_val=1.0).item() 361 | from skimage.metrics import structural_similarity as ssim 362 | # print(out_avg.shape, img.shape) 363 | clean_ssim = ssim(out_avg.detach().cpu().numpy().squeeze(0).squeeze(0), img.detach().cpu().numpy().squeeze(0).squeeze(0)) 364 | ssims_list.append(clean_ssim) 365 | # with torch.no_grad(): 366 | # lpips_score = loss_fn_alex(out_avg.detach().cpu(), img.cpu()).item() 367 | t.set_description("PSNR:{:.5f} db | AVG:{:.5f} | | Loss: {:.5f} | SSIM: {:.5f}".format(clean_psnr, clean_psnr_avg, loss.item(), clean_ssim)) 368 | psnr_list.append(clean_psnr) 369 | loss_list.append(loss.item()) 370 | # lpips_list.append(lpips_score) 371 | # scheduler.step(loss) 372 | 373 | 374 | lpips_list = [0.0] 375 | clean_psnr = psnr(out_avg, img.detach().cpu(), max_val=1.0) 376 | # torch.save(model, '/home/cegrad/calta/sparse-dip/testmodel.pth') 377 | return out_avg, clean_psnr.item(), psnr_list, loss_list, lpips_list, ssims_list 378 | 379 | 380 | 381 | def add_noise(model, param_noise_sigma, learning_rate, dtype): 382 | for n in [x for x in model.parameters() if len(x.size()) == 4]: 383 | noise = torch.randn(n.size())*param_noise_sigma*learning_rate 384 | noise = noise.type(dtype) 385 | n.data = n.data + noise 386 | 387 | --------------------------------------------------------------------------------