AI Paper Atlas

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

AAAI 2026 · Architectures and efficiency › Spiking networks · approach: Event-camera spiking networks · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
use learnable binary sparse encoding instead of EV-VGCNN encoding (ablation)S3Net · N-Caltech101top-1 accuracy (%): 80.80 -> 83.5Improved
use learnable binary sparse encoding instead of VMV-GCN encoding (ablation)S3Net · N-Caltech101top-1 accuracy (%): 80.32 -> 83.5Improved
add motion branch to base branch in the Voxel Spatial Block (dual branch) (ablation)S3Net VSB · N-Caltech101top-1 accuracy (%): 82.3 -> 83.5Improved
add base branch to motion branch in the Voxel Spatial Block (dual branch) (ablation)S3Net VSB · N-Caltech101top-1 accuracy (%): 81.9 -> 83.5Improved
use dual branches with half channels each rather than a single full-channel branch (ablation)S3Net VSB · N-Caltech101parameters (M): 8.49 -> 5.88Improved
use Mamba2 instead of RWKV7 in the Voxel Temporal Block (ablation)S3Net VTB · N-Caltech101top-1 accuracy (%): 81.0 -> 83.5Improved
use Mamba2 instead of LinFormer in the Voxel Temporal Block (ablation)S3Net VTB · N-Caltech101top-1 accuracy (%): 75.8 -> 83.5Improved
sparse voxel design instead of dense frame-based modelsS3Net · N-Caltech101GFlops: 4.28 -> 1.52Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy