Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386072021> ?p ?o ?g. }
- W4386072021 abstract "Generating talking head videos through a face image and a piece of speech audio still contains many challenges. i.e., unnatural head movement, distorted expression, and identity modification. We argue that these issues are mainly caused by learning from the coupled 2D motion fields. On the other hand, explicitly using 3D information also suffers problems of stiff expression and incoherent video. We present SadTalker, which generates 3D motion coefficients (head pose, expression) of the 3DMM from audio and implicitly modulates a novel 3D-aware face render for talking head generation. To learn the realistic motion coefficients, we explicitly model the connections between audio and different types of motion coefficients individually. Precisely, we present ExpNet to learn the accurate facial expression from audio by distilling both coefficients and 3D-rendered faces. As for the head pose, we design PoseVAE via a conditional VAE to synthesize head motion in different styles. Finally, the generated 3D motion coefficients are mapped to the unsupervised 3D keypoints space of the proposed face render to synthesize the final video. We conducted extensive experiments to demonstrate the superiority of our method in terms of motion and video quality. <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>1</sup> <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>1</sup> The code and demo videos are available at https://sadtalker.github.io." @default.
- W4386072021 created "2023-08-23" @default.
- W4386072021 creator A5001967719 @default.
- W4386072021 creator A5007348590 @default.
- W4386072021 creator A5012435102 @default.
- W4386072021 creator A5015030426 @default.
- W4386072021 creator A5043050875 @default.
- W4386072021 creator A5058799911 @default.
- W4386072021 creator A5064222660 @default.
- W4386072021 creator A5086529957 @default.
- W4386072021 date "2023-06-01" @default.
- W4386072021 modified "2023-10-17" @default.
- W4386072021 title "SadTalker: Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation" @default.
- W4386072021 cites W2194775991 @default.
- W4386072021 cites W2237250383 @default.
- W4386072021 cites W2806833697 @default.
- W4386072021 cites W2944294033 @default.
- W4386072021 cites W2962795401 @default.
- W4386072021 cites W2964449965 @default.
- W4386072021 cites W2969985801 @default.
- W4386072021 cites W3019952993 @default.
- W4386072021 cites W3081492798 @default.
- W4386072021 cites W3087121792 @default.
- W4386072021 cites W3174763799 @default.
- W4386072021 cites W3178284600 @default.
- W4386072021 cites W3180391059 @default.
- W4386072021 cites W3180770160 @default.
- W4386072021 cites W3186090335 @default.
- W4386072021 cites W3187364420 @default.
- W4386072021 cites W3195529437 @default.
- W4386072021 cites W3197199219 @default.
- W4386072021 cites W3204680331 @default.
- W4386072021 cites W3211147706 @default.
- W4386072021 cites W4200174933 @default.
- W4386072021 cites W4200630629 @default.
- W4386072021 cites W4200631136 @default.
- W4386072021 cites W4214626920 @default.
- W4386072021 cites W4221145616 @default.
- W4386072021 cites W4281730245 @default.
- W4386072021 cites W4310379947 @default.
- W4386072021 cites W4312301053 @default.
- W4386072021 cites W4312473638 @default.
- W4386072021 cites W4386075576 @default.
- W4386072021 cites W4386076250 @default.
- W4386072021 doi "https://doi.org/10.1109/cvpr52729.2023.00836" @default.
- W4386072021 hasPublicationYear "2023" @default.
- W4386072021 type Work @default.
- W4386072021 citedByCount "0" @default.
- W4386072021 crossrefType "proceedings-article" @default.
- W4386072021 hasAuthorship W4386072021A5001967719 @default.
- W4386072021 hasAuthorship W4386072021A5007348590 @default.
- W4386072021 hasAuthorship W4386072021A5012435102 @default.
- W4386072021 hasAuthorship W4386072021A5015030426 @default.
- W4386072021 hasAuthorship W4386072021A5043050875 @default.
- W4386072021 hasAuthorship W4386072021A5058799911 @default.
- W4386072021 hasAuthorship W4386072021A5064222660 @default.
- W4386072021 hasAuthorship W4386072021A5086529957 @default.
- W4386072021 hasConcept C104114177 @default.
- W4386072021 hasConcept C114793014 @default.
- W4386072021 hasConcept C121684516 @default.
- W4386072021 hasConcept C127313418 @default.
- W4386072021 hasConcept C138591656 @default.
- W4386072021 hasConcept C144024400 @default.
- W4386072021 hasConcept C154945302 @default.
- W4386072021 hasConcept C177264268 @default.
- W4386072021 hasConcept C195704467 @default.
- W4386072021 hasConcept C199360897 @default.
- W4386072021 hasConcept C2776760102 @default.
- W4386072021 hasConcept C2779304628 @default.
- W4386072021 hasConcept C2780312720 @default.
- W4386072021 hasConcept C31972630 @default.
- W4386072021 hasConcept C36289849 @default.
- W4386072021 hasConcept C41008148 @default.
- W4386072021 hasConcept C502989409 @default.
- W4386072021 hasConcept C69369342 @default.
- W4386072021 hasConcept C90559484 @default.
- W4386072021 hasConceptScore W4386072021C104114177 @default.
- W4386072021 hasConceptScore W4386072021C114793014 @default.
- W4386072021 hasConceptScore W4386072021C121684516 @default.
- W4386072021 hasConceptScore W4386072021C127313418 @default.
- W4386072021 hasConceptScore W4386072021C138591656 @default.
- W4386072021 hasConceptScore W4386072021C144024400 @default.
- W4386072021 hasConceptScore W4386072021C154945302 @default.
- W4386072021 hasConceptScore W4386072021C177264268 @default.
- W4386072021 hasConceptScore W4386072021C195704467 @default.
- W4386072021 hasConceptScore W4386072021C199360897 @default.
- W4386072021 hasConceptScore W4386072021C2776760102 @default.
- W4386072021 hasConceptScore W4386072021C2779304628 @default.
- W4386072021 hasConceptScore W4386072021C2780312720 @default.
- W4386072021 hasConceptScore W4386072021C31972630 @default.
- W4386072021 hasConceptScore W4386072021C36289849 @default.
- W4386072021 hasConceptScore W4386072021C41008148 @default.
- W4386072021 hasConceptScore W4386072021C502989409 @default.
- W4386072021 hasConceptScore W4386072021C69369342 @default.
- W4386072021 hasConceptScore W4386072021C90559484 @default.
- W4386072021 hasFunder F4320335777 @default.
- W4386072021 hasLocation W43860720211 @default.
- W4386072021 hasOpenAccess W4386072021 @default.
- W4386072021 hasPrimaryLocation W43860720211 @default.
- W4386072021 hasRelatedWork W1569740590 @default.