Research map / Video / Video generation
Talking Face Generation by Adversarially Disentangled Audio-Visual Representation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add GAN loss to L1 reconstruction (ablation) | DAVS talking-face generation · LRW test, audio reconstruction | PSNR: 25.4 -> 26.7 | Helped |
| add GAN loss to L1 reconstruction (ablation) | DAVS talking-face generation · LRW test, video reconstruction | SSIM: 0.865 -> 0.884 | Helped |
| use audio-visual joint embedding instead of the reproduced baseline | talking-face generation · User study, audio generation | lip-audio sync rate: 58.0% -> 72.3% | Helped |
| add sharing classifier to contrastive-loss training (ablation) | DAVS audio-visual representation · LRW test | visual speech recognition accuracy: 61.8% -> 65.6% | Helped |
| add domain adversarial training to contrastive-loss training (ablation) | DAVS audio-visual representation · LRW test | visual speech recognition accuracy: 61.8% -> 63.5% | Helped |
| combine sharing classifier and domain adversarial training with contrastive loss (ablation) | DAVS audio-visual representation · LRW test | video-to-audio retrieval R@1: 29.3 -> 64.2 | Helped |
| add adversarial disentanglement between identity and speech features (ablation) | DAVS talking-face generation · Audio generation to source | landmark L2: 3.94 -> 3.48 | Helped |
| add adversarial disentanglement between identity and speech features (ablation) | DAVS talking-face generation · Video generation to source | retrieval R@1: 90.8 -> 95.3 | Helped |
Closest papers by meaning
- FaceChain-ImagineID: Freely Crafting High-Fidelity Diverse Talking Faces from Disentangled Audio (CVPR 2024 · similarity 0.90)
- From Faces to Voices: Learning Hierarchical Representations for High-quality Video-to-Speech (CVPR 2025 · similarity 0.89)
- Faces that Speak: Jointly Synthesising Talking Face and Speech from Text (CVPR 2024 · similarity 0.88)
- Perceptually Accurate 3D Talking Head Generation: New Definitions, Speech-Mesh Representation, and Evaluation Metrics (CVPR 2025 · similarity 0.88)
- GoHD: Gaze-oriented and Highly Disentangled Portrait Animation with Rhythmic Poses and Realistic Expressions (AAAI 2025 · similarity 0.88)
- Mimic: Speaking Style Disentanglement for Speech-Driven 3D Facial Animation (AAAI 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.