Research map / Detection and segmentation / CNN features
Online Detection and Classification of Dynamic Hand Gestures With Recurrent 3D Convolutional Neural Network
Check 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 |
|---|---|---|---|
| Fuse all available modalities | R3DCNN · New hand gesture dataset | accuracy: 83.8% | Helped |
| Use depth rather than color input | R3DCNN · New hand gesture dataset | accuracy: 74.1% -> 80.3% | Helped |
| Add a recurrent layer (ablation) | 3D-CNN gesture classifier · New hand gesture dataset, color | accuracy: 67.2% -> 72.0% | Helped |
| Add a recurrent layer (ablation) | 3D-CNN gesture classifier · New hand gesture dataset, depth | accuracy: 73.3% -> 79.5% | Helped |
| Replace 2D-CNN with 3D-CNN (ablation) | Gesture classifier · New hand gesture dataset, color, CTC | accuracy: 65.6% -> 74.1% | Helped |
| Replace 2D-CNN with 3D-CNN (ablation) | Gesture classifier · New hand gesture dataset, depth, CTC | accuracy: 69.1% -> 80.3% | Helped |
| Train with CTC rather than negative log-likelihood (ablation) | 3D-CNN + RNN · New hand gesture dataset, color | accuracy: 72.0% -> 74.1% | Helped |
| Train with CTC rather than negative log-likelihood (ablation) | 3D-CNN + RNN · New hand gesture dataset, depth | accuracy: 79.5% -> 80.3% | Helped |
| Add feature-map dropout (ablation) | 3D-CNN + RNN + CTC · New hand gesture dataset, color | accuracy: 66.5% -> 74.1% | Helped |
| Add feature-map dropout (ablation) | 3D-CNN + RNN + CTC · New hand gesture dataset, depth | accuracy: 75.6% -> 80.3% | Helped |
| Use CTC rather than negative log-likelihood training (ablation) | R3DCNN, depth · New hand gesture dataset, gesture detection | AUC: 0.69 -> 0.91 | Helped |
| Use full-video recurrent features rather than clip-wise C3D features | Linear SVM gesture classifier · New hand gesture dataset, color | accuracy: 69.3% -> 73.0% | Helped |
Closest papers by meaning
- Improving the Performance of Unimodal Dynamic Hand-Gesture Recognition With Multimodal Training (CVPR 2019 · similarity 0.90)
- Regional Attention with Architecture-Rebuilt 3D Network for RGB-D Gesture Recognition (AAAI 2021 · similarity 0.89)
- Gesture Recognition: Focus on the Hands (CVPR 2018 · similarity 0.87)
- Scale Invariant Fully Convolutional Network: Detecting Hands Efficiently (AAAI 2019 · similarity 0.87)
- Personalizing Gesture Recognition Using Hierarchical Bayesian Neural Networks (CVPR 2017 · similarity 0.87)
- Skeleton-Based Gesture Recognition Using Several Fully Connected Layers with Path Signature Features and Temporal Transformer Module (AAAI 2019 · 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.