Research map / Architectures and efficiency / Spiking networks
S3Net: Spatiotemporally Separated Sparse Network for Neuromorphic Vision Processing
Ping He, Rong Xiao, Wanying Xu, Chenwei Tang, Shudong Huang, Huajin Tang
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 |
|---|---|---|---|
| use learnable binary sparse encoding instead of EV-VGCNN encoding (ablation) | S3Net · N-Caltech101 | top-1 accuracy (%): 80.80 -> 83.5 | Improved |
| use learnable binary sparse encoding instead of VMV-GCN encoding (ablation) | S3Net · N-Caltech101 | top-1 accuracy (%): 80.32 -> 83.5 | Improved |
| add motion branch to base branch in the Voxel Spatial Block (dual branch) (ablation) | S3Net VSB · N-Caltech101 | top-1 accuracy (%): 82.3 -> 83.5 | Improved |
| add base branch to motion branch in the Voxel Spatial Block (dual branch) (ablation) | S3Net VSB · N-Caltech101 | top-1 accuracy (%): 81.9 -> 83.5 | Improved |
| use dual branches with half channels each rather than a single full-channel branch (ablation) | S3Net VSB · N-Caltech101 | parameters (M): 8.49 -> 5.88 | Improved |
| use Mamba2 instead of RWKV7 in the Voxel Temporal Block (ablation) | S3Net VTB · N-Caltech101 | top-1 accuracy (%): 81.0 -> 83.5 | Improved |
| use Mamba2 instead of LinFormer in the Voxel Temporal Block (ablation) | S3Net VTB · N-Caltech101 | top-1 accuracy (%): 75.8 -> 83.5 | Improved |
| sparse voxel design instead of dense frame-based models | S3Net · N-Caltech101 | GFlops: 4.28 -> 1.52 | Improved |
Closest papers by meaning
- EventNet: Asynchronous Recursive Event Processing (CVPR 2019 · similarity 0.89)
- Efficient 3D Recognition with Event-driven Spike Sparse Convolution (AAAI 2025 · similarity 0.89)
- Discrete Time Convolution for Fast Event-Based Stereo (CVPR 2022 · similarity 0.89)
- EventZoom: A Progressive Approach to Event-Based Data Augmentation for Enhanced Neuromorphic Vision (AAAI 2025 · similarity 0.88)
- Leveraging Asynchronous Spiking Neural Networks for Ultra Efficient Event-Based Visual Processing (AAAI 2025 · similarity 0.87)
- Efficient Event-Based Object Detection: A Hybrid Neural Network with Spatial and Temporal Attention (CVPR 2025 · 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.
