Research map / Video / Video generation
Streaming Generation of Co-Speech Gestures via Accelerated Rolling Diffusion
Evgeniia Vu, Andrei Boiarov, Dmitry Vetrov
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 |
|---|---|---|---|
| replace seed-pose/outpainting with rolling diffusion framework | DiffuseStyleGesture · ZEGGS | FDg: 6393.99 -> 3831.35 | Improved |
| replace seed-pose/outpainting with rolling diffusion framework | DiffuseStyleGesture · BEAT | FDg: 37062.19 -> 21441.91 | Improved |
| apply rolling diffusion framework | Taming Diffusion · ZEGGS | FDk: 418.85 -> 353.62 | No clear effect |
| apply rolling diffusion framework | DiffSHEG · BEAT | FDg: 13547.4 -> 10201.9 | Improved |
| apply rolling diffusion framework | PersonaGestor · ZEGGS | FDg: 4060.36 -> 3936.75 | Improved |
| add RDLA ladder step l=2 to rolling DSG | DSG rolling · ZEGGS | FDg: 3831.35 -> 5772.40 | Hurt |
| add RDLA ladder step l=4 to rolling DSG | DSG rolling · ZEGGS | FDg: 3831.35 -> 16791.03 | Hurt |
| add RDLA ladder step l=2 to rolling DSG | DSG rolling · BEAT | FDg: 21441.91 -> 17309.63 | Improved |
| add RDLA ladder step l=4 to rolling DSG | DSG rolling · BEAT | Divk: 76.09 -> 83.08 | Improved |
| rolling denoising with fewer steps (1000 -> 100) | DSG rolling · generation speed | FPS: 10 -> 70 | Improved |
| RDLA l=4 vs rolling DSG | DSG · generation speed | FPS: 70 -> 200 | Improved |
| add minimal noise (t=1) to context frames (ablation) | rolling DSG · ZEGGS/BEAT | robustness | Improved |
Closest papers by meaning
- DiffSHEG: A Diffusion-Based Approach for Real-Time Speech-driven Holistic 3D Expression and Gesture Generation (CVPR 2024 · similarity 0.92)
- Co-Speech Gesture Video Generation via Motion-Decoupled Diffusion Model (CVPR 2024 · similarity 0.89)
- ConvoFusion: Multi-Modal Conversational Diffusion for Co-Speech Gesture Synthesis (CVPR 2024 · similarity 0.89)
- Co$^{\mathbf{3}}$Gesture: Towards Coherent Concurrent Co-speech 3D Gesture Generation with Interactive Diffusion (ICLR 2025 · similarity 0.89)
- Co-Speech Gesture Video Generation with Implicit Motion-Audio Entanglement (CVPR 2025 · similarity 0.88)
- DIDiffGes: Decoupled Semi-Implicit Diffusion Models for Real-time Gesture Generation from Speech (AAAI 2025 · 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.
