Research map / Architectures and efficiency / CNN design
Learning Adaptive Hidden Layers for Mobile Gesture Recognition
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 |
|---|---|---|---|
| add selection balancing regularizer (SBR) (ablation) | AHL · IsoGD val RGB+Depth | accuracy: 51.87% -> 54.14% | Helped |
| add selection balancing regularizer (SBR) (ablation) | AHL · collected gesture dataset | accuracy: 90.06% -> 90.57% | Helped |
| replace conventional hidden layers with AHLs (without SBR) (ablation) | multi-modal DAE · collected gesture dataset | accuracy: 86.48% -> 90.06% | Helped |
| replace hidden layers with AHLs (without SBR) (ablation) | C3D+ConvLSTM · IsoGD val RGB+Depth | accuracy: 51.02% -> 51.87% | Helped |
| double neuron count of each layer instead of using AHL (ablation) | multi-modal DAE · collected gesture dataset | accuracy: 86.48% -> 86.02% | No clear effect |
| add one extra conventional hidden layer matching AHL parameters | C3D+ConvLSTM · IsoGD val RGB+Depth | accuracy: 51.02% -> 51.40% | Helped |
| replace last (joint) layer with an AHL (ablation) | multi-modal DAE · collected gesture dataset | error rate: 13.52% -> 11.75% | Helped |
| also replace modality-specific layers with AHLs (ablation) | AHL on joint layer only · collected gesture dataset | error rate: 11.75% -> 9.43% | Helped |
| add multi-modal fusion by concatenating HOG and accelerometer features | DAE · collected gesture dataset | accuracy: 81.52% -> 82.34% | Helped |
| add AHL (RGB+Depth) vs single modality RGB | AHL · IsoGD val | accuracy: 44.88% -> 54.14% | Helped |
Closest papers by meaning
- Online Detection and Classification of Dynamic Hand Gestures With Recurrent 3D Convolutional Neural Network (CVPR 2016 · similarity 0.86)
- SpGesture: Source-Free Domain-adaptive sEMG-based Gesture Recognition with Jaccard Attentive Spiking Neural Network (NeurIPS 2024 · similarity 0.85)
- Revisiting Noise Resilience Strategies in Gesture Recognition: Short-Term Enhancement in sEMG Analysis (ICML 2025 · similarity 0.83)
- SocialGesture: Delving into Multi-person Gesture Understanding (CVPR 2025 · similarity 0.82)
- PixelRNN: In-pixel Recurrent Neural Networks for End-to-end-optimized Perception with Neural Sensors (CVPR 2024 · similarity 0.82)
- PLEIADES: Building Temporal Kernels with Orthogonal Polynomials (NeurIPS 2025 · similarity 0.81)
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.