├── LICENSE ├── README.md └── docs ├── Audio-Driven-Method-Comparsion ├── Arabic_portrait.jpg ├── Arabic_video.mp4 ├── Japanese_portrait.jpg ├── Japanese_video.mp4 ├── Korean_portrait.jpg ├── Korean_video.mp4 ├── Mandarin_portrait.jpg ├── Mandarin_video.mp4 ├── README.txt ├── Swahili_portrait.jpg ├── Swahili_video.mp4 ├── Turkish_portrait.jpg └── Turkish_video.mp4 ├── Coarse-to-fine-visualization ├── 1_first_level_codebook.mp4 ├── 2_first_2_levels_codebook.mp4 ├── 3_first_3_levels_codebook.mp4 ├── 4_all_levels_codebook.mp4 └── README.txt ├── Discrete-Continuous-Representation ├── RD_Radio34_005.mp4_00_RD_Radio34_005.mp4 ├── RD_Radio34_007.mp4_00_RD_Radio34_007.mp4 ├── RD_Radio34_008.mp4_00_RD_Radio34_008.mp4 └── WRA_PeterRoskam0_000.mp4_00_WRA_PeterRoskam0_000.mp4 ├── MNTE-Evaluation-Dataset-Part ├── Arabic.mp4 ├── Japanese.mp4 ├── Korean.mp4 ├── Mandarin.mp4 ├── README.txt ├── Swahili.mp4 └── Turkish.mp4 ├── Multilingual-Audio-Driven-Results ├── Amharic │ ├── 00007572#emop_EP_1#ASED_V1_n1-02-02-02-50.mp4 │ └── 00009603#same_female_mid#ASED_V1_n1-02-01-02-04.mp4 ├── Cantonese │ ├── 00002166#RD_Radio56_000#cosydemo_en_yue.mp4 │ └── 00009013#RD_Radio50_000#cosydemo_jp_yue.mp4 ├── English │ ├── 00000169#3#MEAD_M031_neutral_015.mp4 │ ├── 00000286#15#MEAD_W024_neutral_050.mp4 │ ├── 00002204#WRA_BillCassidy0_000#MEAD_W026_neutral_014.mp4 │ └── 00009494#RD_Radio41_000#MEAD_W029_neutral_021.mp4 ├── French │ ├── 00002918#o6#CaFE_04-N-1.mp4 │ └── 00009198#same_female_right#CaFE_01-N-2.mp4 ├── German │ ├── 00000510#9#EmoDB_15b02Tc.mp4 │ └── 00003600#female_scale_4#EmoDB_16a05Ea.mp4 ├── Italian │ ├── 00001867#male_emotion_1#Emozionalmente_1613823564030.mp4 │ └── 00005397#3#Emozionalmente_1613334970818.mp4 ├── Japanese │ ├── 00003543#female_gaze_1#cosydemo_yue_jp.mp4 │ └── 00007240#o2#cosydemo_en_jp.mp4 ├── Korean │ ├── 00000219#RD_Radio32_000#cosydemo_en_ko.mp4 │ └── 00008885#RD_Radio34_007#cosydemo_zh_ko.mp4 ├── Mandarin │ ├── 00007455#RD_Radio51_000#aishell1_BAC009S0011W0474.mp4 │ └── 00007748#15#aishell1_BAC009S0204W0138.mp4 ├── Turkish │ ├── 00002434#17#TurEV_1234_sk_sucuk.mp4 │ └── 00003781#17#TurEV_1234_sk_demet.mp4 └── URDU │ ├── 00001304#15#URDU_SM5_F5_N033.mp4 │ └── 00007679#male_emotion_4#URDU_SM5_F38_N066.mp4 ├── README.txt ├── VQ-Ablation-Studies ├── Cross_WRA_BillCassidy0_000.mp4_00_WRA_LamarAlexander0_000.mp4 ├── RD_Radio34_005.mp4_00_RD_Radio34_005.mp4 ├── RD_Radio34_007.mp4_00_RD_Radio34_007.mp4 ├── RD_Radio35_000.mp4_00_RD_Radio35_000.mp4 ├── WRA_LamarAlexander0_000.mp4_00_WRA_LamarAlexander0_000.mp4 └── WRA_LamarAlexander_000.mp4_00_WRA_LamarAlexander_000.mp4 ├── Video-Reconstruction ├── RD_Radio32_000.mp4_00_RD_Radio32_000.mp4 ├── RD_Radio34_007.mp4_00_RD_Radio34_007.mp4 ├── RD_Radio50_000.mp4_00_RD_Radio50_000.mp4 └── RD_Radio56_000.mp4_00_RD_Radio56_000.mp4 └── index.html /LICENSE: -------------------------------------------------------------------------------- 1 | Apache License 2 | Version 2.0, January 2004 3 | http://www.apache.org/licenses/ 4 | 5 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 6 | 7 | 1. Definitions. 8 | 9 | "License" shall mean the terms and conditions for use, reproduction, 10 | and distribution as defined by Sections 1 through 9 of this document. 11 | 12 | "Licensor" shall mean the copyright owner or entity authorized by 13 | the copyright owner that is granting the License. 14 | 15 | "Legal Entity" shall mean the union of the acting entity and all 16 | other entities that control, are controlled by, or are under common 17 | control with that entity. For the purposes of this definition, 18 | "control" means (i) the power, direct or indirect, to cause the 19 | direction or management of such entity, whether by contract or 20 | otherwise, or (ii) ownership of fifty percent (50%) or more of the 21 | outstanding shares, or (iii) beneficial ownership of such entity. 22 | 23 | "You" (or "Your") shall mean an individual or Legal Entity 24 | exercising permissions granted by this License. 25 | 26 | "Source" form shall mean the preferred form for making modifications, 27 | including but not limited to software source code, documentation 28 | source, and configuration files. 29 | 30 | "Object" form shall mean any form resulting from mechanical 31 | transformation or translation of a Source form, including but 32 | not limited to compiled object code, generated documentation, 33 | and conversions to other media types. 34 | 35 | "Work" shall mean the work of authorship, whether in Source or 36 | Object form, made available under the License, as indicated by a 37 | copyright notice that is included in or attached to the work 38 | (an example is provided in the Appendix below). 39 | 40 | "Derivative Works" shall mean any work, whether in Source or Object 41 | form, that is based on (or derived from) the Work and for which the 42 | editorial revisions, annotations, elaborations, or other modifications 43 | represent, as a whole, an original work of authorship. For the purposes 44 | of this License, Derivative Works shall not include works that remain 45 | separable from, or merely link (or bind by name) to the interfaces of, 46 | the Work and Derivative Works thereof. 47 | 48 | "Contribution" shall mean any work of authorship, including 49 | the original version of the Work and any modifications or additions 50 | to that Work or Derivative Works thereof, that is intentionally 51 | submitted to Licensor for inclusion in the Work by the copyright owner 52 | or by an individual or Legal Entity authorized to submit on behalf of 53 | the copyright owner. For the purposes of this definition, "submitted" 54 | means any form of electronic, verbal, or written communication sent 55 | to the Licensor or its representatives, including but not limited to 56 | communication on electronic mailing lists, source code control systems, 57 | and issue tracking systems that are managed by, or on behalf of, the 58 | Licensor for the purpose of discussing and improving the Work, but 59 | excluding communication that is conspicuously marked or otherwise 60 | designated in writing by the copyright owner as "Not a Contribution." 61 | 62 | "Contributor" shall mean Licensor and any individual or Legal Entity 63 | on behalf of whom a Contribution has been received by Licensor and 64 | subsequently incorporated within the Work. 65 | 66 | 2. Grant of Copyright License. Subject to the terms and conditions of 67 | this License, each Contributor hereby grants to You a perpetual, 68 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 69 | copyright license to reproduce, prepare Derivative Works of, 70 | publicly display, publicly perform, sublicense, and distribute the 71 | Work and such Derivative Works in Source or Object form. 72 | 73 | 3. Grant of Patent License. Subject to the terms and conditions of 74 | this License, each Contributor hereby grants to You a perpetual, 75 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 76 | (except as stated in this section) patent license to make, have made, 77 | use, offer to sell, sell, import, and otherwise transfer the Work, 78 | where such license applies only to those patent claims licensable 79 | by such Contributor that are necessarily infringed by their 80 | Contribution(s) alone or by combination of their Contribution(s) 81 | with the Work to which such Contribution(s) was submitted. If You 82 | institute patent litigation against any entity (including a 83 | cross-claim or counterclaim in a lawsuit) alleging that the Work 84 | or a Contribution incorporated within the Work constitutes direct 85 | or contributory patent infringement, then any patent licenses 86 | granted to You under this License for that Work shall terminate 87 | as of the date such litigation is filed. 88 | 89 | 4. Redistribution. You may reproduce and distribute copies of the 90 | Work or Derivative Works thereof in any medium, with or without 91 | modifications, and in Source or Object form, provided that You 92 | meet the following conditions: 93 | 94 | (a) You must give any other recipients of the Work or 95 | Derivative Works a copy of this License; and 96 | 97 | (b) You must cause any modified files to carry prominent notices 98 | stating that You changed the files; and 99 | 100 | (c) You must retain, in the Source form of any Derivative Works 101 | that You distribute, all copyright, patent, trademark, and 102 | attribution notices from the Source form of the Work, 103 | excluding those notices that do not pertain to any part of 104 | the Derivative Works; and 105 | 106 | (d) If the Work includes a "NOTICE" text file as part of its 107 | distribution, then any Derivative Works that You distribute must 108 | include a readable copy of the attribution notices contained 109 | within such NOTICE file, excluding those notices that do not 110 | pertain to any part of the Derivative Works, in at least one 111 | of the following places: within a NOTICE text file distributed 112 | as part of the Derivative Works; within the Source form or 113 | documentation, if provided along with the Derivative Works; or, 114 | within a display generated by the Derivative Works, if and 115 | wherever such third-party notices normally appear. The contents 116 | of the NOTICE file are for informational purposes only and 117 | do not modify the License. You may add Your own attribution 118 | notices within Derivative Works that You distribute, alongside 119 | or as an addendum to the NOTICE text from the Work, provided 120 | that such additional attribution notices cannot be construed 121 | as modifying the License. 122 | 123 | You may add Your own copyright statement to Your modifications and 124 | may provide additional or different license terms and conditions 125 | for use, reproduction, or distribution of Your modifications, or 126 | for any such Derivative Works as a whole, provided Your use, 127 | reproduction, and distribution of the Work otherwise complies with 128 | the conditions stated in this License. 129 | 130 | 5. Submission of Contributions. Unless You explicitly state otherwise, 131 | any Contribution intentionally submitted for inclusion in the Work 132 | by You to the Licensor shall be under the terms and conditions of 133 | this License, without any additional terms or conditions. 134 | Notwithstanding the above, nothing herein shall supersede or modify 135 | the terms of any separate license agreement you may have executed 136 | with Licensor regarding such Contributions. 137 | 138 | 6. Trademarks. This License does not grant permission to use the trade 139 | names, trademarks, service marks, or product names of the Licensor, 140 | except as required for reasonable and customary use in describing the 141 | origin of the Work and reproducing the content of the NOTICE file. 142 | 143 | 7. Disclaimer of Warranty. Unless required by applicable law or 144 | agreed to in writing, Licensor provides the Work (and each 145 | Contributor provides its Contributions) on an "AS IS" BASIS, 146 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or 147 | implied, including, without limitation, any warranties or conditions 148 | of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A 149 | PARTICULAR PURPOSE. You are solely responsible for determining the 150 | appropriateness of using or redistributing the Work and assume any 151 | risks associated with Your exercise of permissions under this License. 152 | 153 | 8. Limitation of Liability. In no event and under no legal theory, 154 | whether in tort (including negligence), contract, or otherwise, 155 | unless required by applicable law (such as deliberate and grossly 156 | negligent acts) or agreed to in writing, shall any Contributor be 157 | liable to You for damages, including any direct, indirect, special, 158 | incidental, or consequential damages of any character arising as a 159 | result of this License or out of the use or inability to use the 160 | Work (including but not limited to damages for loss of goodwill, 161 | work stoppage, computer failure or malfunction, or any and all 162 | other commercial damages or losses), even if such Contributor 163 | has been advised of the possibility of such damages. 164 | 165 | 9. Accepting Warranty or Additional Liability. While redistributing 166 | the Work or Derivative Works thereof, You may choose to offer, 167 | and charge a fee for, acceptance of support, warranty, indemnity, 168 | or other liability obligations and/or rights consistent with this 169 | License. However, in accepting such obligations, You may act only 170 | on Your own behalf and on Your sole responsibility, not on behalf 171 | of any other Contributor, and only if You agree to indemnify, 172 | defend, and hold each Contributor harmless for any liability 173 | incurred by, or claims asserted against, such Contributor by reason 174 | of your accepting any such warranty or additional liability. 175 | 176 | END OF TERMS AND CONDITIONS 177 | 178 | APPENDIX: How to apply the Apache License to your work. 179 | 180 | To apply the Apache License to your work, attach the following 181 | boilerplate notice, with the fields enclosed by brackets "[]" 182 | replaced with your own identifying information. (Don't include 183 | the brackets!) The text should be enclosed in the appropriate 184 | comment syntax for the file format. We also recommend that a 185 | file or class name and description of purpose be included on the 186 | same "printed page" as the copyright notice for easier 187 | identification within third-party archives. 188 | 189 | Copyright [yyyy] [name of copyright owner] 190 | 191 | Licensed under the Apache License, Version 2.0 (the "License"); 192 | you may not use this file except in compliance with the License. 193 | You may obtain a copy of the License at 194 | 195 | http://www.apache.org/licenses/LICENSE-2.0 196 | 197 | Unless required by applicable law or agreed to in writing, software 198 | distributed under the License is distributed on an "AS IS" BASIS, 199 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 200 | See the License for the specific language governing permissions and 201 | limitations under the License. 202 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | 2 | 3 | # VQTalker 4 | 5 | Towards Multilingual Talking Avatars through Facial Motion Tokenization 6 | 7 | 8 |
9 | 10 | 11 | 12 |
13 |
14 | 15 | Updating 16 | 17 | -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Arabic_portrait.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Arabic_portrait.jpg -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Arabic_video.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Arabic_video.mp4 -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Japanese_portrait.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Japanese_portrait.jpg -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Japanese_video.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Japanese_video.mp4 -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Korean_portrait.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Korean_portrait.jpg -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Korean_video.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Korean_video.mp4 -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Mandarin_portrait.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Mandarin_portrait.jpg -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Mandarin_video.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Mandarin_video.mp4 -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/README.txt: -------------------------------------------------------------------------------- 1 | This folder contains comparison videos of our method and other methods. The pronunciation demonstrated is the same as in Figure 4, to illustrate the advantages of our algorithm for non-Indo-European languages. 2 | 3 | JPG files with the "_portrait" suffix are single portraits that need to be animated (sourced from: https://thispersondoesnotexist.com/) 4 | 5 | MP4 files with the "_video" suffix are the algorithm comparison results. From left to right, they show: SadTalker, EAT, PD-FGC, AniTalker, EDTalker, EchoMimic, VQTalker (Ours) 6 | 7 | All video are resized to a unified resolution: 256*256. -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Swahili_portrait.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Swahili_portrait.jpg -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Swahili_video.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Swahili_video.mp4 -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Turkish_portrait.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Turkish_portrait.jpg -------------------------------------------------------------------------------- /docs/Audio-Driven-Method-Comparsion/Turkish_video.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Audio-Driven-Method-Comparsion/Turkish_video.mp4 -------------------------------------------------------------------------------- /docs/Coarse-to-fine-visualization/1_first_level_codebook.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Coarse-to-fine-visualization/1_first_level_codebook.mp4 -------------------------------------------------------------------------------- /docs/Coarse-to-fine-visualization/2_first_2_levels_codebook.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Coarse-to-fine-visualization/2_first_2_levels_codebook.mp4 -------------------------------------------------------------------------------- /docs/Coarse-to-fine-visualization/3_first_3_levels_codebook.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Coarse-to-fine-visualization/3_first_3_levels_codebook.mp4 -------------------------------------------------------------------------------- /docs/Coarse-to-fine-visualization/4_all_levels_codebook.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Coarse-to-fine-visualization/4_all_levels_codebook.mp4 -------------------------------------------------------------------------------- /docs/Coarse-to-fine-visualization/README.txt: -------------------------------------------------------------------------------- 1 | This folder contains a set of four videos demonstrating the progressive impact of residual codebook levels in our video driving model. Each video is divided into three panels: the leftmost displays the source portrait to be driven, the middle shows the driving video, and the rightmost presents the driven result. To illustrate the modeling capabilities at different residual codebook levels, we sequentially reveal the codebooks: 2 | 3 | 1. "1_first_level_codebook.mp4" showcases only the first codebook level, masking the subsequent three. 4 | 2. "2_first_2_levels_codebook.mp4" reveals the first two codebook levels, masking the latter two. 5 | 3. "3_first_3_levels_codebook.mp4" utilizes the first three codebook levels, masking only the final one. 6 | 4. "4_all_levels_codebook.mp4" employs all codebook levels. 7 | 8 | Observing the progression, we find that the first codebook level models coarse-grained features such as head pose incompletely. Introducing the second residual codebook level captures finer details like eye blinks and lip movements. The third codebook level achieves a near-complete reconstruction of the original video, though some jitter remains. Finally, incorporating all codebook levels (the complete set) eliminates the remaining jitter, producing a smooth and accurate result. This demonstration underscores the hierarchical nature of our residual codebook approach, showcasing how each level contributes to increasingly refined and faithful video reconstruction. -------------------------------------------------------------------------------- /docs/Discrete-Continuous-Representation/RD_Radio34_005.mp4_00_RD_Radio34_005.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Discrete-Continuous-Representation/RD_Radio34_005.mp4_00_RD_Radio34_005.mp4 -------------------------------------------------------------------------------- /docs/Discrete-Continuous-Representation/RD_Radio34_007.mp4_00_RD_Radio34_007.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Discrete-Continuous-Representation/RD_Radio34_007.mp4_00_RD_Radio34_007.mp4 -------------------------------------------------------------------------------- /docs/Discrete-Continuous-Representation/RD_Radio34_008.mp4_00_RD_Radio34_008.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Discrete-Continuous-Representation/RD_Radio34_008.mp4_00_RD_Radio34_008.mp4 -------------------------------------------------------------------------------- /docs/Discrete-Continuous-Representation/WRA_PeterRoskam0_000.mp4_00_WRA_PeterRoskam0_000.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Discrete-Continuous-Representation/WRA_PeterRoskam0_000.mp4_00_WRA_PeterRoskam0_000.mp4 -------------------------------------------------------------------------------- /docs/MNTE-Evaluation-Dataset-Part/Arabic.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/MNTE-Evaluation-Dataset-Part/Arabic.mp4 -------------------------------------------------------------------------------- /docs/MNTE-Evaluation-Dataset-Part/Japanese.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/MNTE-Evaluation-Dataset-Part/Japanese.mp4 -------------------------------------------------------------------------------- /docs/MNTE-Evaluation-Dataset-Part/Korean.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/MNTE-Evaluation-Dataset-Part/Korean.mp4 -------------------------------------------------------------------------------- /docs/MNTE-Evaluation-Dataset-Part/Mandarin.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/MNTE-Evaluation-Dataset-Part/Mandarin.mp4 -------------------------------------------------------------------------------- /docs/MNTE-Evaluation-Dataset-Part/README.txt: -------------------------------------------------------------------------------- 1 | Due to capacity limitations of supplementary materials, this directory contains only a portion of the Multilingual Non-Indo-European Talking Head Evaluation Corpus (MNTE) dataset files. In the future, we will at least make public the public web sources. After we ensure that there are no copyright issues, we will release all the data for everyone to test. -------------------------------------------------------------------------------- /docs/MNTE-Evaluation-Dataset-Part/Swahili.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/MNTE-Evaluation-Dataset-Part/Swahili.mp4 -------------------------------------------------------------------------------- /docs/MNTE-Evaluation-Dataset-Part/Turkish.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/MNTE-Evaluation-Dataset-Part/Turkish.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Amharic/00007572#emop_EP_1#ASED_V1_n1-02-02-02-50.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Amharic/00007572#emop_EP_1#ASED_V1_n1-02-02-02-50.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Amharic/00009603#same_female_mid#ASED_V1_n1-02-01-02-04.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Amharic/00009603#same_female_mid#ASED_V1_n1-02-01-02-04.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Cantonese/00002166#RD_Radio56_000#cosydemo_en_yue.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Cantonese/00002166#RD_Radio56_000#cosydemo_en_yue.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Cantonese/00009013#RD_Radio50_000#cosydemo_jp_yue.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Cantonese/00009013#RD_Radio50_000#cosydemo_jp_yue.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/English/00000169#3#MEAD_M031_neutral_015.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/English/00000169#3#MEAD_M031_neutral_015.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/English/00000286#15#MEAD_W024_neutral_050.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/English/00000286#15#MEAD_W024_neutral_050.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/English/00002204#WRA_BillCassidy0_000#MEAD_W026_neutral_014.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/English/00002204#WRA_BillCassidy0_000#MEAD_W026_neutral_014.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/English/00009494#RD_Radio41_000#MEAD_W029_neutral_021.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/English/00009494#RD_Radio41_000#MEAD_W029_neutral_021.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/French/00002918#o6#CaFE_04-N-1.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/French/00002918#o6#CaFE_04-N-1.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/French/00009198#same_female_right#CaFE_01-N-2.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/French/00009198#same_female_right#CaFE_01-N-2.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/German/00000510#9#EmoDB_15b02Tc.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/German/00000510#9#EmoDB_15b02Tc.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/German/00003600#female_scale_4#EmoDB_16a05Ea.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/German/00003600#female_scale_4#EmoDB_16a05Ea.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Italian/00001867#male_emotion_1#Emozionalmente_1613823564030.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Italian/00001867#male_emotion_1#Emozionalmente_1613823564030.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Italian/00005397#3#Emozionalmente_1613334970818.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Italian/00005397#3#Emozionalmente_1613334970818.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Japanese/00003543#female_gaze_1#cosydemo_yue_jp.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Japanese/00003543#female_gaze_1#cosydemo_yue_jp.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Japanese/00007240#o2#cosydemo_en_jp.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Japanese/00007240#o2#cosydemo_en_jp.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Korean/00000219#RD_Radio32_000#cosydemo_en_ko.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Korean/00000219#RD_Radio32_000#cosydemo_en_ko.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Korean/00008885#RD_Radio34_007#cosydemo_zh_ko.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Korean/00008885#RD_Radio34_007#cosydemo_zh_ko.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Mandarin/00007455#RD_Radio51_000#aishell1_BAC009S0011W0474.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Mandarin/00007455#RD_Radio51_000#aishell1_BAC009S0011W0474.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Mandarin/00007748#15#aishell1_BAC009S0204W0138.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Mandarin/00007748#15#aishell1_BAC009S0204W0138.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Turkish/00002434#17#TurEV_1234_sk_sucuk.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Turkish/00002434#17#TurEV_1234_sk_sucuk.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/Turkish/00003781#17#TurEV_1234_sk_demet.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/Turkish/00003781#17#TurEV_1234_sk_demet.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/URDU/00001304#15#URDU_SM5_F5_N033.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/URDU/00001304#15#URDU_SM5_F5_N033.mp4 -------------------------------------------------------------------------------- /docs/Multilingual-Audio-Driven-Results/URDU/00007679#male_emotion_4#URDU_SM5_F38_N066.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Multilingual-Audio-Driven-Results/URDU/00007679#male_emotion_4#URDU_SM5_F38_N066.mp4 -------------------------------------------------------------------------------- /docs/README.txt: -------------------------------------------------------------------------------- 1 | This includes the Supplementary Materials for VQTalker. 2 | 3 | - index.html 4 | Please open it directly. It contains visual files of the videos, including the following visual materials: 5 | 6 | 1. Audio-driven Talking Face Generation (Multi-lingual) 7 | 2. Audio-driven Method Comparison 8 | 3. Video Reconstruction 9 | 4. Coarse-to-fine Visualization 10 | 5. Codebook Ablation 11 | 6. Discrete vs. Continuous Representation 12 | 7. MNTE-Evaluation-Dataset 13 | 14 | If you encounter any issues opening `index.html`, you can also directly check the demo in the corresponding folder. -------------------------------------------------------------------------------- /docs/VQ-Ablation-Studies/Cross_WRA_BillCassidy0_000.mp4_00_WRA_LamarAlexander0_000.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/VQ-Ablation-Studies/Cross_WRA_BillCassidy0_000.mp4_00_WRA_LamarAlexander0_000.mp4 -------------------------------------------------------------------------------- /docs/VQ-Ablation-Studies/RD_Radio34_005.mp4_00_RD_Radio34_005.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/VQ-Ablation-Studies/RD_Radio34_005.mp4_00_RD_Radio34_005.mp4 -------------------------------------------------------------------------------- /docs/VQ-Ablation-Studies/RD_Radio34_007.mp4_00_RD_Radio34_007.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/VQ-Ablation-Studies/RD_Radio34_007.mp4_00_RD_Radio34_007.mp4 -------------------------------------------------------------------------------- /docs/VQ-Ablation-Studies/RD_Radio35_000.mp4_00_RD_Radio35_000.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/VQ-Ablation-Studies/RD_Radio35_000.mp4_00_RD_Radio35_000.mp4 -------------------------------------------------------------------------------- /docs/VQ-Ablation-Studies/WRA_LamarAlexander0_000.mp4_00_WRA_LamarAlexander0_000.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/VQ-Ablation-Studies/WRA_LamarAlexander0_000.mp4_00_WRA_LamarAlexander0_000.mp4 -------------------------------------------------------------------------------- /docs/VQ-Ablation-Studies/WRA_LamarAlexander_000.mp4_00_WRA_LamarAlexander_000.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/VQ-Ablation-Studies/WRA_LamarAlexander_000.mp4_00_WRA_LamarAlexander_000.mp4 -------------------------------------------------------------------------------- /docs/Video-Reconstruction/RD_Radio32_000.mp4_00_RD_Radio32_000.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Video-Reconstruction/RD_Radio32_000.mp4_00_RD_Radio32_000.mp4 -------------------------------------------------------------------------------- /docs/Video-Reconstruction/RD_Radio34_007.mp4_00_RD_Radio34_007.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Video-Reconstruction/RD_Radio34_007.mp4_00_RD_Radio34_007.mp4 -------------------------------------------------------------------------------- /docs/Video-Reconstruction/RD_Radio50_000.mp4_00_RD_Radio50_000.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Video-Reconstruction/RD_Radio50_000.mp4_00_RD_Radio50_000.mp4 -------------------------------------------------------------------------------- /docs/Video-Reconstruction/RD_Radio56_000.mp4_00_RD_Radio56_000.mp4: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/X-LANCE/VQTalker/140d36031986d1f12d77e7dd2f4b3a8617141707/docs/Video-Reconstruction/RD_Radio56_000.mp4_00_RD_Radio56_000.mp4 -------------------------------------------------------------------------------- /docs/index.html: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | VQ-Talker 7 | 8 | 9 | 10 | 28 | 29 | 30 | 31 |
32 |
33 |

34 | VQTalker: Towards Multilingual Talking Avatars through Facial Motion Tokenization 35 |

36 |

37 | Demos 38 | | 39 | Repo 40 |

41 |
42 |
43 | 44 |
45 |
46 |

Index Links:

47 | 58 | 59 |
60 |
61 | 62 | 63 |
64 |
65 |
66 |
67 |
68 |

Audio-driven Talking Face Generation (Multi-lingual)

69 |
70 |

71 | Note: 72 |

    73 |
  • This section demonstrates the robustness of the algorithm in comparing multiple languages.
    • 74 |
  • The left side shows the driven portrait, while the right side displays the images generated by our model, with poses taken from random videos.
    • 75 |
  • All faces in this section are fictional and do not exist. The images of people are randomly picked from thispersondoesnotexist website. We express our special thanks to this website.
    • 76 |
77 | 78 |

79 |
80 |

English

81 |

82 | 83 | 84 | 85 | 86 | 87 | 88 | 89 | 90 | 91 | 92 | 93 | 94 | 95 | 96 |
97 |

Audio samples are from MEAD

98 |
99 | 100 | 101 |

102 |

Below are the generalized results for unseen languages:

103 |

Amharic

104 |

105 | 106 | 107 | 108 | 109 | 110 | 111 | 112 | 113 | 114 |
115 |

Audio samples are from ASED

116 |
117 | 118 |

119 |

Cantonese

120 |

121 | 122 | 123 | 124 | 125 | 126 | 127 | 128 | 129 |
130 |

Audio samples are from CosyVoice

131 |
132 | 133 | 134 |

135 |

French

136 |

137 | 138 | 139 | 140 | 141 | 142 | 143 | 144 |
145 |

Audio samples are from CaFE

146 | 147 | 148 |
149 |

150 |

German

151 |

152 | 153 | 154 | 155 | 156 | 157 | 158 | 159 |
160 |

Audio samples are from EmoDB

161 |
162 | 163 |

Italian

164 |

165 | 166 | 167 | 168 | 169 | 170 | 171 | 172 |
173 |

Audio samples are from Emozionalmente

174 | 175 |
176 | 177 |

Japanese

178 |

179 | 180 | 181 | 182 | 183 | 184 | 185 | 186 |
187 |

Audio samples are from CosyVoice

188 |
189 | 190 |

191 |

Korean

192 |

193 | 194 | 195 | 196 | 197 | 198 | 199 | 200 |
201 |

Audio samples are from CosyVoice

202 |
203 | 204 | 205 |

206 |

Mandarin

207 |

208 | 209 | 210 | 211 | 212 | 213 | 214 | 215 |
216 |

Audio samples are from Aishell1

217 | 218 |
219 | 220 | 221 | 222 | 223 | 224 |

225 |

Turkish

226 |

227 | 228 | 229 | 230 | 231 | 232 | 233 | 234 |
235 |

Audio samples are from TurEV-DB

236 | 237 |
238 | 239 | 240 |

241 |

URDU

242 |

243 | 244 | 245 | 246 | 247 | 248 | 249 | 250 |
251 |

Audio samples are from URDU

252 | 253 |
254 |
255 |
256 |
257 | 258 |
259 |
260 |
261 |
262 |

Audio-driven Method Comparison

263 |
264 |

265 | Note: 266 |

    267 |
  • This section demonstrates the comparison between our method and other methods in audio-driven scenarios.
    • 268 |
  • The source portrait is the driven frames, and below are the generated results.
    • 269 |
  • If these algorithms can control posture, blinking, etc., these control signals come from the source video.
    • 270 |
271 |

272 |
273 |

Arabic (MNTE dataset)

274 |

275 | 276 | 277 | 278 | 281 | 282 | 283 | 284 | 287 | 288 | 289 | 290 | 293 | 294 | 295 | 296 | 310 | 311 | 312 | 313 |
279 |

Source Portrait:

280 |
285 | Portrait Image 286 |
291 |

Result:

292 |
297 | 300 |
301 | SadTalker 302 | EAT 303 | PD-FGC 304 | AniTalker 305 | EDTalker 306 | EchoMimic 307 | VQTalker (Ours) 308 |
309 |
314 |
315 |
316 |
317 |
318 | 319 |
320 |
321 |
322 |
323 | 324 |

Japanese (MNTE dataset)

325 |

326 | 327 | 328 | 329 | 332 | 333 | 334 | 335 | 338 | 339 | 340 | 341 | 344 | 345 | 346 | 347 | 361 | 362 | 363 | 364 |
330 |

Source Portrait:

331 |
336 | Portrait Image 337 |
342 |

Result:

343 |
348 | 351 |
352 | SadTalker 353 | EAT 354 | PD-FGC 355 | AniTalker 356 | EDTalker 357 | EchoMimic 358 | VQTalker (Ours) 359 |
360 |
365 |
366 |
367 |
368 |
369 | 370 | 371 |
372 |
373 |
374 |
375 | 376 |

Korean (MNTE dataset)

377 |

378 | 379 | 380 | 381 | 384 | 385 | 386 | 387 | 390 | 391 | 392 | 393 | 396 | 397 | 398 | 399 | 413 | 414 | 415 | 416 |
382 |

Source Portrait:

383 |
388 | Portrait Image 389 |
394 |

Result:

395 |
400 | 403 |
404 | SadTalker 405 | EAT 406 | PD-FGC 407 | AniTalker 408 | EDTalker 409 | EchoMimic 410 | VQTalker (Ours) 411 |
412 |
417 |
418 |
419 |
420 |
421 | 422 | 423 |
424 |
425 |
426 |
427 | 428 |

Mandarin (MNTE dataset)

429 |

430 | 431 | 432 | 433 | 436 | 437 | 438 | 439 | 442 | 443 | 444 | 445 | 448 | 449 | 450 | 451 | 465 | 466 | 467 | 468 |
434 |

Source Portrait:

435 |
440 | Portrait Image 441 |
446 |

Result:

447 |
452 | 455 |
456 | SadTalker 457 | EAT 458 | PD-FGC 459 | AniTalker 460 | EDTalker 461 | EchoMimic 462 | VQTalker (Ours) 463 |
464 |
469 |
470 |
471 |
472 |
473 | 474 | 475 |
476 |
477 |
478 |
479 | 480 |

Swahili (MNTE dataset)

481 |

482 | 483 | 484 | 485 | 488 | 489 | 490 | 491 | 494 | 495 | 496 | 497 | 500 | 501 | 502 | 503 | 517 | 518 | 519 | 520 |
486 |

Source Portrait:

487 |
492 | Portrait Image 493 |
498 |

Result:

499 |
504 | 507 |
508 | SadTalker 509 | EAT 510 | PD-FGC 511 | AniTalker 512 | EDTalker 513 | EchoMimic 514 | VQTalker (Ours) 515 |
516 |
521 |
522 |
523 |
524 |
525 | 526 |
527 |
528 |
529 |
530 | 531 |

Turkish (MNTE dataset)

532 |

533 | 534 | 535 | 536 | 539 | 540 | 541 | 542 | 545 | 546 | 547 | 548 | 551 | 552 | 553 | 554 | 568 | 569 | 570 | 571 |
537 |

Source Portrait:

538 |
543 | Portrait Image 544 |
549 |

Result:

550 |
555 | 558 |
559 | SadTalker 560 | EAT 561 | PD-FGC 562 | AniTalker 563 | EDTalker 564 | EchoMimic 565 | VQTalker (Ours) 566 |
567 |
572 |
573 |
574 |
575 |
576 |
577 | 578 |
579 |
580 |
581 |
582 |
583 |

Video Reconstruction

584 |
585 |

586 | Note: 587 |

    588 |
  • This section demonstrates the reconstruction capability of our face tokenizer
    • 589 |
  • We use the first frame of the video and subsequent features (extracted from subsequent frames) to reconstruct the entire video
    • 590 |
  • This video highlights our compact and efficient features, with the lowest bitrate among these methods, approximately 11kbps
    • 591 |
592 |

593 |
594 | 595 | 596 | 597 | 598 | 601 | 602 | 603 | 604 | 621 | 622 | 623 |
599 |

Video1:

600 |
605 | 608 |
609 | H.264
(347 kbps) 610 | FOMM
(48 kbps) 611 | DPE
(16 kbps) 612 | MTIA
(48 kbps) 613 | Vid2Vid
(36 kbps) 614 | LIA
(16 kbps) 615 | FADM
(36 kbps) 616 | AniTalke
(16 kbps) 617 | LivePortrait
(50 kbps) 618 | FaceTokenizer
(Ours, 11 kbps) 619 |
620 |
624 |
625 |
626 |
627 |
628 |
629 |
630 |
631 | 632 | 633 | 634 | 635 | 638 | 639 | 640 | 641 | 658 | 659 | 660 |
636 |

Video2:

637 |
642 | 645 |
646 | H.264
(347 kbps) 647 | FOMM
(48 kbps) 648 | DPE
(16 kbps) 649 | MTIA
(48 kbps) 650 | Vid2Vid
(36 kbps) 651 | LIA
(16 kbps) 652 | FADM
(36 kbps) 653 | AniTalke
(16 kbps) 654 | LivePortrait
(50 kbps) 655 | FaceTokenizer
(Ours, 11 kbps) 656 |
657 |
661 |
662 |
663 |
664 |
665 |
666 |
667 |
668 | 669 | 670 | 671 | 672 | 675 | 676 | 677 | 678 | 695 | 696 | 697 |
673 |

Video3:

674 |
679 | 682 |
683 | H.264
(347 kbps) 684 | FOMM
(48 kbps) 685 | DPE
(16 kbps) 686 | MTIA
(48 kbps) 687 | Vid2Vid
(36 kbps) 688 | LIA
(16 kbps) 689 | FADM
(36 kbps) 690 | AniTalke
(16 kbps) 691 | LivePortrait
(50 kbps) 692 | FaceTokenizer
(Ours, 11 kbps) 693 |
694 |
698 |
699 |
700 |
701 |
702 |
703 |
704 |
705 | 706 | 707 | 708 | 709 | 712 | 713 | 714 | 715 | 732 | 733 | 734 |
710 |

Video4:

711 |
716 | 719 |
720 | H.264
(347 kbps) 721 | FOMM
(48 kbps) 722 | DPE
(16 kbps) 723 | MTIA
(48 kbps) 724 | Vid2Vid
(36 kbps) 725 | LIA
(16 kbps) 726 | FADM
(36 kbps) 727 | AniTalke
(16 kbps) 728 | LivePortrait
(50 kbps) 729 | FaceTokenizer
(Ours, 11 kbps) 730 |
731 |
735 |
736 |
737 |
738 |
739 |
740 | 741 | 742 |
743 |
744 |
745 |
746 |
747 |

Coarse-to-fine Visualization

748 |
749 |

750 | Note: 751 |

    752 |
  • This section demonstrates the modeling capabilities of different residual layers, and we will mask some residual layers.
    • 753 |
  • We use the first frame of the video and subsequent features (extracted from VQ features) to reconstruct the entire video.
    • 754 |
  • This video highlights the generation of residual VQ from coarse to fine granularity.
    • 755 |
756 |

757 |
758 |

759 | Observing the progression, we find that the first codebook level models coarse-grained features such as head pose incompletely. Introducing the second residual codebook level captures finer details like eye blinks and lip movements. The third codebook level achieves a near-complete reconstruction of the original video, though some jitter remains. Finally, incorporating all codebook levels (the complete set) eliminates the remaining jitter, producing a smooth and accurate result. This demonstration underscores the hierarchical nature of our residual codebook approach, showcasing how each level contributes to increasingly refined and faithful video reconstruction. 760 |

761 |
762 |
763 | 764 | 765 | 766 | 767 | 770 | 771 | 772 | 773 | 783 | 784 | 785 | 786 |
768 |

(1) only the first codebook level, masking the subsequent three:

769 |
774 | 777 |
778 | Source
 779 | Driven Feature
(Providing VQ features for driving) 780 | Result
 781 |
782 |
787 |
788 |
789 |
790 |
791 |
792 |
793 |
794 | 795 | 796 | 797 | 798 | 801 | 802 | 803 | 804 | 814 | 815 | 816 | 817 |
799 |

(2) the first two codebook levels, masking the latter two:

800 |
805 | 808 |
809 | Source
 810 | Driven Feature
(Providing VQ features for driving) 811 | Result
 812 |
813 |
818 |
819 |
820 |
821 |
822 |
823 |
824 |
825 | 826 | 827 | 828 | 829 | 832 | 833 | 834 | 835 | 845 | 846 | 847 | 848 |
830 |

(3) the first three codebook levels, masking only the final one:

831 |
836 | 839 |
840 | Source
 841 | Driven Feature
(Providing VQ features for driving) 842 | Result
 843 |
844 |
849 |
850 |
851 |
852 |
853 |
854 |
855 |
856 | 857 | 858 | 859 | 860 | 863 | 864 | 865 | 866 | 876 | 877 | 878 | 879 |
861 |

(4) all codebook levels:

862 |
867 | 870 |
871 | Source
 872 | Driven Feature
(Providing VQ features for driving) 873 | Result
 874 |
875 |
880 |
881 |
882 | 883 |
884 | 885 | 886 |
887 | 888 |
889 |
890 |
891 |
892 |
893 |

Codebook Ablation

894 |
895 |

896 | Note: 897 |

    898 |
  • We tried different VQ methods as the bottleneck structure of our face tokenizer
    • 899 |
  • The order of the methods in the figure below is consistent with the order in the paper's chart
    • 900 |
901 |

902 |
903 |
904 |
905 | 906 | 907 | 908 | 909 | 912 | 913 | 914 | 915 | 931 | 932 | 933 | 934 |
910 |

Video 1:

911 |
916 | 919 |
920 | Driven Video
 921 | VQ
 922 | GVQ
 923 | RVQ
 924 | GRVQ
 925 | GRFSQ #1
 926 | GRFSQ #2
 927 | GRFSQ #3
 928 | GRFSQ #4
(Ours)
929 | 930 |
935 |
936 |
937 |
938 |
939 | 940 | 941 | 942 | 943 | 946 | 947 | 948 | 949 | 965 | 966 | 967 | 968 |
944 |

Video 2:

945 |
950 | 953 |
954 | Driven Video
 955 | VQ
 956 | GVQ
 957 | RVQ
 958 | GRVQ
 959 | GRFSQ #1
 960 | GRFSQ #2
 961 | GRFSQ #3
 962 | GRFSQ #4
(Ours)
963 | 964 |
969 |
970 |
971 |
972 |
973 | 974 | 975 | 976 | 977 | 980 | 981 | 982 | 983 | 999 | 1000 | 1001 | 1002 |
978 |

Video 3:

979 |
984 | 987 |
988 | Driven Video
 989 | VQ
 990 | GVQ
 991 | RVQ
 992 | GRVQ
 993 | GRFSQ #1
 994 | GRFSQ #2
 995 | GRFSQ #3
 996 | GRFSQ #4
(Ours)
997 | 998 |
1003 |
1004 | 1005 |
1006 |
1007 | 1008 | 1009 | 1010 | 1011 | 1014 | 1015 | 1016 | 1017 | 1033 | 1034 | 1035 | 1036 |
1012 |

Video 4:

1013 |
1018 | 1021 |
1022 | Driven Video
 1023 | VQ
 1024 | GVQ
 1025 | RVQ
 1026 | GRVQ
 1027 | GRFSQ #1
 1028 | GRFSQ #2
 1029 | GRFSQ #3
 1030 | GRFSQ #4
(Ours)
1031 | 1032 |
1037 |
1038 |
1039 |
1040 | 1041 | 1042 | 1043 | 1044 | 1047 | 1048 | 1049 | 1050 | 1066 | 1067 | 1068 | 1069 |
1045 |

Video 5 (Cross Video Driven):

1046 |
1051 | 1054 |
1055 | Driven Video
 1056 | VQ
 1057 | GVQ
 1058 | RVQ
 1059 | GRVQ
 1060 | GRFSQ #1
 1061 | GRFSQ #2
 1062 | GRFSQ #3
 1063 | GRFSQ #4
(Ours)
1064 | 1065 |
1070 |
1071 | 1072 |
1073 | 1074 |
1075 |
1076 |
1077 |
1078 |
1079 |

Discrete vs. Continuous Representation

1080 |
1081 |

1082 | Note: 1083 |

    1084 |
  • We tested different types of features as input and output
    • 1085 |
  • C represents continuous features, and D represents discrete features
    • 1086 |
  • The configuration of the methods here is consistent with the order of the table in the paper
    • 1087 |
1088 |

1089 |
1090 | 1091 | 1092 | 1093 | 1094 | 1097 | 1098 | 1099 | 1100 | 1112 | 1113 | 1114 | 1115 |
1095 |

Video 1:

1096 |
1101 | 1104 |
1105 | C-to-C
(Whisper Continuous Vector to Continuous Vector) 1106 | D-to-C
(CosyVoice Speech Tokens to Continuous Vector) 1107 | C-to-D
(Whisper Continuous Vector to Discrete Vector) 1108 | D-to-D
(VQ-wav2vec Speech Tokens to Discrete Vector) 1109 | D-to-D
(CosyVoice Speech Tokens to Discrete Vector)
(Ours) 1110 |
1111 |
1116 |
1117 |
1118 |
1119 |
1120 |
1121 |
1122 | 1123 | 1124 | 1125 | 1126 | 1129 | 1130 | 1131 | 1132 | 1144 | 1145 | 1146 | 1147 |
1127 |

Video 2:

1128 |
1133 | 1136 |
1137 | C-to-C
(Whisper Continuous Vector to Continuous Vector) 1138 | D-to-C
(CosyVoice Speech Tokens to Continuous Vector) 1139 | C-to-D
(Whisper Continuous Vector to Discrete Vector) 1140 | D-to-D
(VQ-wav2vec Speech Tokens to Discrete Vector) 1141 | D-to-D
(CosyVoice Speech Tokens to Discrete Vector)
(Ours) 1142 |
1143 |
1148 |
1149 |
1150 | 1151 |
1152 |
1153 |
1154 | 1155 | 1156 | 1157 | 1158 | 1161 | 1162 | 1163 | 1164 | 1176 | 1177 | 1178 | 1179 |
1159 |

Video 3:

1160 |
1165 | 1168 |
1169 | C-to-C
(Whisper Continuous Vector to Continuous Vector) 1170 | D-to-C
(CosyVoice Speech Tokens to Continuous Vector) 1171 | C-to-D
(Whisper Continuous Vector to Discrete Vector) 1172 | D-to-D
(VQ-wav2vec Speech Tokens to Discrete Vector) 1173 | D-to-D
(CosyVoice Speech Tokens to Discrete Vector)
(Ours) 1174 |
1175 |
1180 |
1181 |
1182 | 1183 |
1184 |
1185 |
1186 | 1187 | 1188 | 1189 | 1190 | 1193 | 1194 | 1195 | 1196 | 1208 | 1209 | 1210 | 1211 |
1191 |

Video 4:

1192 |
1197 | 1200 |
1201 | C-to-C
(Whisper Continuous Vector to Continuous Vector) 1202 | D-to-C
(CosyVoice Speech Tokens to Continuous Vector) 1203 | C-to-D
(Whisper Continuous Vector to Discrete Vector) 1204 | D-to-D
(VQ-wav2vec Speech Tokens to Discrete Vector) 1205 | D-to-D
(CosyVoice Speech Tokens to Discrete Vector)
(Ours) 1206 |
1207 |
1212 |
1213 |
1214 | 1215 |
1216 | 1217 |
1218 |
1219 |
1220 |
1221 |
1222 |

MNTE Evaluation Dataset

1223 |
1224 |

1225 | Due to capacity limitations of supplementary materials, this directory contains only a portion of the Multilingual Non-Indo-European Talking Head Evaluation Corpus (MNTE) dataset files. In the future, we will at least publish the source of the videos, so you can download them yourself. After we ensure that there are no copyright issues, we will release all the data for everyone to test. 1226 |

1227 |
1228 |
1229 | 1230 | 1231 | 1232 | 1233 | 1234 | 1235 | 1236 | 1237 | 1238 | 1239 | 1240 |
1241 |
1242 | Arabic
 1243 | Japanese
 1244 | Korean
 1245 | Mandarin
 1246 | Swahili
 1247 | Turkish
 1248 |
1249 |
1250 |
1251 |
1252 |
1253 |
1254 |
1255 |
1256 |

1257 | Ethical Consideration 1258 |

1259 |
1260 |

1261 | The rapid advancement of digital human technology, particularly in the creation of highly realistic virtual faces, presents significant ethical challenges. There are genuine concerns about the potential misuse of this technology for malicious purposes, such as deepfakes, identity theft, or the propagation of misinformation. To address these issues, it is crucial that developers and organizations establish comprehensive ethical guidelines before deploying such technologies. These guidelines should encompass principles of user privacy, data protection, and responsible use. Furthermore, to enhance accountability and prevent misuse, it is recommended to implement robust verification systems and content attribution methods for all digitally generated human representations. This could include blockchain-based authentication or secure metadata tagging. By proactively addressing these ethical considerations, we can foster the positive potential of digital human technology while minimizing its risks to individuals and society. 1262 |

1263 |
1264 |
1265 |
1266 | 1267 | 1268 | 1294 | 1295 | 1296 | --------------------------------------------------------------------------------