Research map / Image generation / Autoregressive and latent generation
QPGesture: Quantization-Based and Phase-Guided Motion Matching for Natural Speech-Driven Gesture Generation
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 |
|---|---|---|---|
| Use vq-wav2vec audio quantization and Levenshtein distance instead of WavLM and cosine similarity (ablation) | QPGesture · BEAT test set | FGD on raw data space: 6009.859 -> 5742.281 | Helped |
| Include audio modality alongside text (ablation) | QPGesture · BEAT test set | FGD on feature space: 20.401 -> 19.921 | Helped |
| Include text modality alongside audio (ablation) | QPGesture · BEAT test set | FGD on feature space: 23.929 -> 19.921 | Helped |
| Add phase guidance to candidate selection (ablation) | QPGesture · BEAT test set | Human-likeness: 3.90 ± 0.11 -> 4.07 ± 0.15 | No clear effect |
| Use motion matching instead of a GRU to select gesture codes (ablation) | QPGesture · BEAT test set | FGD on feature space: 30.404 -> 19.921 | Helped |
Closest papers by meaning
- DiffSHEG: A Diffusion-Based Approach for Real-Time Speech-driven Holistic 3D Expression and Gesture Generation (CVPR 2024 · similarity 0.87)
- Towards Variable and Coordinated Holistic Co-Speech Motion Generation (CVPR 2024 · similarity 0.87)
- Co$^{\mathbf{3}}$Gesture: Towards Coherent Concurrent Co-speech 3D Gesture Generation with Interactive Diffusion (ICLR 2025 · similarity 0.87)
- ConvoFusion: Multi-Modal Conversational Diffusion for Co-Speech Gesture Synthesis (CVPR 2024 · similarity 0.87)
- Co-Speech Gesture Video Generation with Implicit Motion-Audio Entanglement (CVPR 2025 · similarity 0.87)
- Co-Speech Gesture Video Generation via Motion-Decoupled Diffusion Model (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.