Research map / Detection and segmentation / CNN features
Scale Invariant Fully Convolutional Network: Detecting Hands Efficiently
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 multi-scale loss to SIFCN with VGG16+UF (ablation) | SIFCN (VGG16+UF) · Oxford (bigger hands) | AP (%): 68.7 -> 77.8 | Helped |
| Add multi-scale loss to SIFCN with ResNet50+UF (ablation) | SIFCN (ResNet50+UF) · Oxford (bigger hands) | AP (%): 78.2 -> 78.6 | Helped |
| Replace UF with CWF (ablation) | SIFCN with VGG16 and multi-scale loss · Oxford (bigger hands) | AP (%): 77.8 -> 78.0 | Helped |
| Replace UF with CWF (ablation) | SIFCN with ResNet50 and multi-scale loss · Oxford (bigger hands) | AP (%): 78.6 -> 80.4 | Helped |
| Add multi-scale loss supervision at intermediate layers (ablation) | SIFCN · Oxford | training time to convergence: more than 10 hours | Helped |
| Add multi-scale loss supervision at intermediate layers (ablation) | SIFCN · Oxford | AP | No clear effect |
| Increase score-map loss weight from 0.01 to 0.10 (ablation) | SIFCN · Oxford | AP: 0.01 to 1; maximum when α is 0.10 | Helped |
| Include rotation-angle loss (β > 0) rather than setting β = 0 (ablation) | SIFCN · Oxford | AP: larger than 0.78; 0.4991 | Helped |
| Replace UF with CWF (ablation) | SIFCN · VIVA Level-1 and Level-2 | AR | Helped |
| Use ResNet50 instead of VGG16 (ablation) | SIFCN · Oxford (bigger hands) | AP (%): 80.4 vs. 78.0 | Helped |
Closest papers by meaning
- Detecting Hands and Recognizing Physical Contact in the Wild (NeurIPS 2020 · similarity 0.89)
- Harmonious Feature Learning for Interactive Hand-Object Pose Estimation (CVPR 2023 · similarity 0.88)
- Online Detection and Classification of Dynamic Hand Gestures With Recurrent 3D Convolutional Neural Network (CVPR 2016 · similarity 0.87)
- Hand PointNet: 3D Hand Pose Estimation Using Point Sets (CVPR 2018 · similarity 0.86)
- SiMHand: Mining Similar Hands for Large-Scale 3D Hand Pose Pre-training (ICLR 2025 · similarity 0.86)
- Learning Dense Hand Contact Estimation from Imbalanced Data (NeurIPS 2025 · similarity 0.84)
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.