Research map / Video / Video generation
From Noise to Events: Conditional Diffusion for Event Data Augmentation
Ruofei Wang, Ziyuan Luo, Peiqi Duan, Xiufeng HUANG, Boxin Shi, Renjie Wan
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 |
|---|---|---|---|
| augment with N2E | Swin, Voxel Grid · N-Caltech101 | top-1 accuracy: 91.3 -> 94.9 | Improved |
| augment with N2E | ViT, Voxel Grid · DVS-Gesture | top-1 accuracy: 63.9 -> 83.7 | Improved |
| augment with N2E | Swin, Voxel Grid · DVS-Gesture | top-1 accuracy: 74.3 -> 93.4 | Improved |
| augment with EventMix | ResNet18, Event Frame · N-Caltech101 | top-1 accuracy: 73.7 -> 53.0 | Hurt |
| augment with EventMix | VGG, Event Frame · N-Caltech101 | top-1 accuracy: 64.4 -> 72.8 | Improved |
| augment with DataAug | ResNet18, Event Frame · DVS-Gesture | top-1 accuracy: 77.8 -> 85.4 | Improved |
| augment with DataAug | ResNet18, Event Spike Tensor · DVS-Gesture | top-1 accuracy: 86.1 -> 89.6 | Improved |
| augment with EventDrop | Time Surface models · DVS-Gesture | training convergence | Hurt |
| augment with EventDrop | ViT and VGG · N-Caltech101 | top-1 accuracy | Hurt |
| augment with ShapeAug | ResNet50, Event Spike Tensor · DVS-Gesture | top-1 accuracy: 0.4% improvement | No clear effect |
| augment with ShapeAug | five models · N-Caltech101 | top-1 accuracy | No clear effect |
| use N2E instead of baseline augmentation | event data quality · MVSEC | ePSNR: 2.1627 -> 11.9747 | Improved |
Closest papers by meaning
- EventZoom: A Progressive Approach to Event-Based Data Augmentation for Enhanced Neuromorphic Vision (AAAI 2025 · similarity 0.88)
- E-Motion: Future Motion Simulation via Event Sequence Diffusion (NeurIPS 2024 · similarity 0.86)
- LaSe-E2V: Towards Language-guided Semantic-aware Event-to-Video Reconstruction (NeurIPS 2024 · similarity 0.85)
- V2V: Scaling Event-Based Vision through Efficient Video-to-Voxel Simulation (NeurIPS 2025 · similarity 0.85)
- EventDance: Unsupervised Source-free Cross-modal Adaptation for Event-based Object Recognition (CVPR 2024 · similarity 0.85)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 2025 · similarity 0.85)
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.
