Research map / Graph learning / Knowledge and temporal graphs
Graph-Based Asynchronous Event Processing for Rapid Object Recognition
Yijin Li, Han Zhou, Bangbang Yang, Ye Zhang, Zhaopeng Cui, Hujun Bao, Guofeng Zhang
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 |
|---|---|---|---|
| replace graph convolution with slide convolution (ablation) | EvS · N-Caltech101 | computation per event (Mps/ev): 1152 -> 11.5 | Improved |
| replace graph convolution with slide convolution (ablation) | EvS · CIFAR10-DVS | computation per event (Mps/ev): 3020 -> 33.2 | Improved |
| replace graph convolution with slide convolution (ablation) | EvS · N-Caltech101 | accuracy: 0.761 -> 0.761 | No clear effect |
| replace graph convolution with slide convolution (ablation) | NVS · N-Caltech101 | computation per event (Mps/ev): 221 -> 7.8 | Improved |
| replace graph convolution with slide convolution (ablation) | EvS · N-Caltech101 | processing time per event update: 130.4 ms -> 16.9 ms | Improved |
| replace k-d tree (nanoflann) with pixel-queue radius search (ablation) | graph construction · synthetic window of 100,000 events | search time: reduces the time cost by half | Improved |
| replace k-d tree (nanoflann) with pixel-queue radius search (ablation) | graph construction · synthetic window of 100,000 events | insertion and deletion cost: an order of magnitude lower | Improved |
| use event-wise SlideGCN instead of batch-wise 50 ms processing | EvS · N-Caltech101 (stream) | accuracy: 0.761 -> 0.669 | Hurt |
| use event-wise SlideGCN instead of batch-wise 50 ms processing | EvS · N-Caltech101 (stream) | event index at response: 19154 -> 3118 | Improved |
| reduce batch size from 50 ms to 10 ms (ablation) | batch-wise EvS · N-Caltech101 (stream) | accuracy: 0.761 -> 0.362 | Hurt |
| increase mini-batch size from 1 to 100 (ablation) | SlideGCN · N-Caltech101 (stream) | cumulative MFLOPS: 17711 -> 3041 | Improved |
| increase mini-batch size from 1 to 100 (ablation) | SlideGCN · N-Caltech101 (stream) | average MFLOPS: 5.68 -> 95.64 | Hurt |
Closest papers by meaning
- Graph Neural Network Combining Event Stream and Periodic Aggregation for Low-Latency Event-based Vision (CVPR 2025 · similarity 0.89)
- EGSST: Event-based Graph Spatiotemporal Sensitive Transformer for Object Detection (NeurIPS 2024 · similarity 0.87)
- Asynchronous Collaborative Graph Representation for Frames and Events (CVPR 2025 · similarity 0.87)
- Leveraging Asynchronous Spiking Neural Networks for Ultra Efficient Event-Based Visual Processing (AAAI 2025 · similarity 0.87)
- EventPillars: Pillar-based Efficient Representations for Event Data (AAAI 2025 · similarity 0.86)
- A Dynamic GCN with Cross-Representation Distillation for Event-Based Learning (AAAI 2024 · 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.
