Research map / Video / Video generation
MimicTalker: A Multimodal Interactive and Memory-Enhanced Framework for Real-Time Dyadic 3D Head Generation
Yinuo Wang, Yanbo Fan, Xuan Wang, Boyao Zhou, Yu Guo, Yujun Shen, Fei Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 MICE (contextually-aware memory of interlocutor) (ablation) | MimicTalker baseline · DualTalk test | FD expression (x10^3): 7.95 -> 7.45 | Improved |
| add SDI (intention and topic semantics) (ablation) | MimicTalker baseline · DualTalk test | FD expression (x10^3): 7.95 -> 7.69 | Improved |
| add MSM (motion style memory) (ablation) | MimicTalker baseline · DualTalk test | FD expression (x10^3): 7.95 -> 7.39 | Improved |
| add all three modules (ablation) | MimicTalker baseline · DualTalk test | FD expression (x10^3): 7.95 -> 7.12 | Improved |
| add MSM on out-of-distribution clips (ablation) | MimicTalker baseline · DualTalk OOD | FD expression (x10^3): 19.39 -> 16.17 | Improved |
| add MICE (ablation) | MimicTalker baseline · DualTalk test | rPCC expression (x10^3): 4.58 -> 4.24 | Improved |
| add MSM on long dialogues (ablation) | MimicTalker baseline · Seamless Interaction | FD expression (x10^3): 27.28 -> 25.58 | Improved |
| add all modules on long dialogues (ablation) | MimicTalker baseline · Seamless Interaction | FD expression (x10^3): 27.28 -> 24.52 | Improved |
| add MICE or SDI or MSM to baseline (ablation) | MimicTalker baseline · DualTalk test | SID expression (x10^3, diversity): 3.51 -> 3.58 / 3.56 / 3.59 | No clear effect |
| use MimicTalker instead of DualTalk | MimicTalker · DualTalk test | FD expression (x10^3): 11.14 -> 7.12 | Improved |
Closest papers by meaning
- INFP: Audio-Driven Interactive Head Generation in Dyadic Conversations (CVPR 2025 · similarity 0.91)
- DualTalk: Dual-Speaker Interaction for 3D Talking Head Conversations (CVPR 2025 · similarity 0.88)
- Motion-Agent: A Conversational Framework for Human Motion Generation with LLMs (ICLR 2025 · similarity 0.86)
- AnyTalk: Multi-modal Driven Multi-domain Talking Head Generation (AAAI 2025 · similarity 0.86)
- Full-Duplex Speech-Motion Model for Dyadic Interaction (NeurIPS 2026 · similarity 0.86)
- FaceTalk: Audio-Driven Motion Diffusion for Neural Parametric Head Models (CVPR 2024 · similarity 0.86)
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.
