Research map / Detection and segmentation / CNN features
Adaptive Spatial-Temporal Window: Unlocking the Potential of Event Cameras in Heterogeneous Velocity Scenarios
Zhipeng Sui, Haiqing Hao, Weihua He, Seng-Hong Lee, Wenhui 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 patch division (spatial locality) (ablation) | ASTW · Gen1 detection, ResNet-50 | mAP: 46.6 -> 50.6 | Improved |
| add causal consistency (minimum time-step strategy) (ablation) | ASTW · Gen1 detection, ResNet-50 | mAP: 49.4 -> 50.6 | Improved |
| add spatial smoothing of event density (ablation) | ASTW · Gen1 detection, ResNet-50 | mAP: 50.1 -> 50.6 | Improved |
| add temporal EMA smoothing of event density (ablation) | ASTW · Gen1 detection, ResNet-50 | mAP: 50.3 -> 50.6 | Improved |
| replace Fixed Time Window with ASTW, Time Surface representation | Faster R-CNN ResNet-50 · Gen1 detection | mAP: 51.3 -> 53.0 | Improved |
| replace Fixed Time Window with ASTW, Event Count representation | Faster R-CNN ResNet-50 · Gen1 detection | mAP: 50.1 -> 52.1 | Improved |
| replace Fixed Time Window with ASTW, Voxel Grid representation | Faster R-CNN ResNet-50 · Gen1 detection | mAP: 47.6 -> 49.0 | Improved |
| replace Fixed Time Window with ASTW on recurrent RVT | RVT · Gen1 detection | mAP: 46.9 -> 48.3 | Improved |
| replace Fixed Time Window with ASTW for tracking on EventVOT | HDETrack · EventVOT | SR: 61.7 -> 62.3 | Improved |
| replace Fixed Time Window with ASTW on HetVel | HDETrack · HetVel | SR: 47.7 -> 50.7 | Improved |
| use patch size 4 instead of 64 (ablation) | ASTW · Gen1 detection, ResNet-50 | mAP: 48.9 -> 50.6 | Improved |
| vary information constant gamma from 1.0 to 2.0 (ablation) | ASTW · Gen1 detection, ResNet-50 | mAP: 50.4 -> 50.6 | No clear effect |
Closest papers by meaning
- EventZoom: A Progressive Approach to Event-Based Data Augmentation for Enhanced Neuromorphic Vision (AAAI 2025 · similarity 0.89)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 2025 · similarity 0.89)
- Event Stream-based Visual Object Tracking: A High-Resolution Benchmark Dataset and A Novel Baseline (CVPR 2024 · similarity 0.88)
- FlexEvent: Towards Flexible Event-Frame Object Detection at Varying Operational Frequencies (NeurIPS 2025 · similarity 0.88)
- TimeTracker: Event-based Continuous Point Tracking for Video Frame Interpolation with Non-linear Motion (CVPR 2025 · similarity 0.87)
- Bridge Frame and Event: Common Spatiotemporal Fusion for High-Dynamic Scene Optical Flow (CVPR 2025 · similarity 0.87)
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.
