Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
BiCrossNet with Decoupled Dual Generators: A Parameter‑Efficient and Generalizable Few‑Shot Custom Gesture Recognition Framework
Chunyang Yu, Ning SUN, Shenyue Wang
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 data augmentation (Embedding Generator) (ablation) | BiCroNet · UMAHand | F1 (%): 54.56 -> 98.13 | Improved |
| add data augmentation (Embedding Generator) (ablation) | BiCroNet · GRHand | F1 (%): 54.51 -> 96.13 | Improved |
| replace Delta Encoder with Delta Generator (ablation) | BiCroNet · UMAHand | F1 (%): 92.50 -> 97.14 | Improved |
| replace Delta Encoder with Embedding Generator (ablation) | BiCroNet · UMAHand | F1 (%): 92.50 -> 98.13 | Improved |
| replace Delta Encoder with Delta Generator (ablation) | BiCroNet · GRHand | F1 (%): 89.51 -> 90.99 | Improved |
| replace Delta Encoder with Embedding Generator (ablation) | BiCroNet · GRHand | F1 (%): 89.51 -> 96.13 | Improved |
| replace Delta Encoder with Delta Generator (ablation) | Improved TCNN · UMAHand | accuracy (%): 82.18 -> 97.39 | Improved |
| replace Delta Encoder with Embedding Generator (ablation) | Improved TCNN · UMAHand | accuracy (%): 82.18 -> 98.10 | Improved |
| replace Delta Encoder with Embedding Generator (ablation) | Improved TCNN · GRHand | F1 (%): 81.04 -> 77.99 | Mixed |
| use BiCrossNet extractor instead of Improved TCNN (Embedding Generator) (ablation) | Embedding Generator · UMAHand | parameters: 45198 -> 32642 | Improved |
| use BiCrossNet extractor instead of AppleDE (Delta Encoder) | Delta Encoder · UMAHand | F1 (%): 88.31 -> 92.50 | Improved |
| use BiCrossNet extractor instead of Apple extractor | CPU inference · Xeon Gold 6342 | latency (ms): 11.07 -> 9.16 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Delta Encoder — Delta-encoder: an effective sample synthesis method for few-shot object recognition (NeurIPS 2018) · 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
- Widar3.0: Zero-Effort Cross-Domain Gesture Recognition with Wi-Fi (TPAMI 2021 · similarity 0.84)
- SpGesture: Source-Free Domain-adaptive sEMG-based Gesture Recognition with Jaccard Attentive Spiking Neural Network (NeurIPS 2024 · similarity 0.83)
- Learning Adaptive Hidden Layers for Mobile Gesture Recognition (AAAI 2018 · similarity 0.83)
- GigaHands: A Massive Annotated Dataset of Bimanual Hand Activities (CVPR 2025 · similarity 0.82)
- Revisiting Noise Resilience Strategies in Gesture Recognition: Short-Term Enhancement in sEMG Analysis (ICML 2025 · similarity 0.81)
- Hand1000: Generating Realistic Hands from Text with Only 1,000 Images (AAAI 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.
