Research map / Video / Video generation
SemGes: Semantics-aware Co-Speech Gesture Generation using Semantic Coherence and Relevance Learning
Lanmiao Liu, Esam Ghaleb, Asli Ozyurek, Zerrin Yumak
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 semantic coherence loss (ablation) | SemGes · BEAT | FGD: 8.053 -> 4.467 | Improved |
| add semantic coherence loss (ablation) | SemGes · BEAT | SRGR: 0.180 -> 0.256 | Improved |
| add semantic coherence loss (ablation) | SemGes · BEAT | Diversity: 249.550 -> 305.706 | Improved |
| add semantic relevance loss (ablation) | SemGes · BEAT | FGD: 7.549 -> 4.467 | Improved |
| add semantic relevance loss (ablation) | SemGes · BEAT | SRGR: 0.195 -> 0.256 | Improved |
| add semantic coherence loss (ablation) | SemGes · TED-Expressive | FGD: 7.924 -> 7.263 | Improved |
| add semantic coherence loss (ablation) | SemGes · TED-Expressive | Diversity: 109.256 -> 302.772 | Improved |
| use HuBERT speech encoder instead of SpeechCLIP (ablation) | SemGes · BEAT | FGD: 6.787 -> 4.467 | Improved |
| use HuBERT speech encoder instead of SpeechCLIP (ablation) | SemGes · TED-Expressive | FGD: 7.341 -> 7.263 | Improved |
| add stage 2 (two-stage design) over VQ-VAE baseline (ablation) | SemGes · BEAT | FGD: 10.348 -> 4.467 | Improved |
| use SemGes instead of strongest baseline DiffSheg | SemGes · BEAT | BC: 0.922 -> 0.453 | Hurt |
| evaluate on beat-dominant gestures without semantic content | SemGes · BEAT | BC: 0.453 -> 0.689 | Improved |
Later papers that compared with this method
2 later accepted papers report SemGes as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- CtrlCoMo: Controllable Co-Speech Motion Generation with Gesture–Action Disentanglement (ECCV 2026 · better on 3 of 3 shared results)
- StreamTalk: Streaming Co-Speech Gesture Generation with Key-Pose Anchoring (ECCV 2026 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Co$^{\mathbf{3}}$Gesture: Towards Coherent Concurrent Co-speech 3D Gesture Generation with Interactive Diffusion (ICLR 2025 · similarity 0.91)
- Retrieving Semantics from the Deep: an RAG Solution for Gesture Synthesis (CVPR 2025 · similarity 0.90)
- ConvoFusion: Multi-Modal Conversational Diffusion for Co-Speech Gesture Synthesis (CVPR 2024 · similarity 0.90)
- HOP: Heterogeneous Topology-based Multimodal Entanglement for Co-Speech Gesture Generation (CVPR 2025 · similarity 0.90)
- Co-Speech Gesture Video Generation with Implicit Motion-Audio Entanglement (CVPR 2025 · similarity 0.89)
- Towards Variable and Coordinated Holistic Co-Speech Motion Generation (CVPR 2024 · similarity 0.89)
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.
