Research map / Vision-language and multimodal / Cross-modal learning
Deep Dynamic Neural Networks for Multimodal Gesture Segmentation and Recognition
Di Wu, Lionel Pigou, Pieter-Jan Kindermans, Nam Le, Ling Shao, Joni Dambre, Jean‐Marc Odobez
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 RGB-D 3DCNN module to skeleton DBN via late fusion (ablation) | Skeleton DBN · ChaLearn LAP gesture spotting (test) | Jaccard index: 0.779 -> 0.809 | Improved |
| add skeleton module to RGB-D 3DCNN via late fusion (ablation) | RGB-D 3DCNN · ChaLearn LAP gesture spotting (test) | Jaccard index: 0.717 -> 0.809 | Improved |
| add intermediate fusion of skeleton and RGB-D (ablation) | Skeleton DBN · ChaLearn LAP gesture spotting (test) | Jaccard index: 0.779 -> 0.798 | Improved |
| use intermediate fusion instead of late fusion | Multimodal DDNN · ChaLearn LAP gesture spotting (test) | Jaccard index: 0.809 -> 0.798 | Hurt |
| add HMM temporal model with Viterbi decoding (ablation) | Late fusion DDNN · ChaLearn LAP gesture spotting (test) | Jaccard index: 0:66 -> 0.809 | Improved |
| add HMM temporal model with Viterbi decoding (ablation) | Late fusion DDNN · ChaLearn LAP gesture spotting event level (test) | Recognized (%): 76:6 -> 86.4 | Improved |
| add skeleton module to RGB-D module (late fusion) (ablation) | RGB-D 3DCNN · ChaLearn LAP gesture spotting event level (test) | Missed (%): 19.7 -> 4.9 | Improved |
| add RGB-D module to skeleton module (late fusion) (ablation) | Skeleton DBN · ChaLearn LAP gesture spotting event level (test) | Confused (%): 12.3 -> 8.7 | Improved |
| use 10 states per gesture instead of 5 (ablation) | ES-HMM · ChaLearn LAP gesture spotting | Jaccard index | No clear effect |
Closest papers by meaning
- Online Detection and Classification of Dynamic Hand Gestures With Recurrent 3D Convolutional Neural Network (CVPR 2016 · similarity 0.90)
- Improving the Performance of Unimodal Dynamic Hand-Gesture Recognition With Multimodal Training (CVPR 2019 · similarity 0.88)
- MMTM: Multimodal Transfer Module for CNN Fusion (CVPR 2020 · similarity 0.86)
- S2Gest: Split-Scan State Space Models for Dynamic Hand Gesture Recognition (ECCV 2026 · similarity 0.85)
- Regional Attention with Architecture-Rebuilt 3D Network for RGB-D Gesture Recognition (AAAI 2021 · similarity 0.85)
- Beyond Temporal Pooling: Recurrence and Temporal Convolutions for Gesture Recognition in Video (IJCV 2016 · similarity 0.85)
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.
