Research map / Video / Video segmentation and tracking
NER-Net+: Seeing Motion at Nighttime With an Event Camera
Haoyue Liu, Jinghan Xu, Shihan Peng, Yi Chang, Hanyu Zhou, Yuxing Duan, Lin Zhu, Yonghong Tian, Luxin Yan
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 |
|---|---|---|---|
| feed ETS-processed events directly without LETC (ablation) | base reconstruction network · RLED | LPIPS: 0.393 -> 0.423 | Hurt |
| add LETC (learnable timestamp calibration) (ablation) | base reconstruction network · RLED | LPIPS: 0.393 -> 0.368 | Improved |
| add ETS together with LETC (ablation) | LETC · RLED | LPIPS: 0.368 -> 0.354 | Improved |
| add Local Adaptation Gate (LAG) (ablation) | ETS+LETC · RLED | LPIPS: 0.354 -> 0.331 | Improved |
| add GCB (ablation) | ETS+LETC+LAG · RLED | LPIPS: 0.331 -> 0.324 | Mixed |
| add SAU (completing NIAM) (ablation) | ETS+LETC+LAG+GCB · RLED | LPIPS: 0.324 -> 0.309 | Improved |
| train E2VID+ on V2E simulated events instead of RLED (ablation) | E2VID+ · RLED | MSE: 0.019 -> 0.100 (MSE) | Hurt |
| train NER-Net on RLED instead of V2E simulated events (ablation) | NER-Net · RLED | MSE: 0.087 -> 0.011 (MSE) | Improved |
| train NER-Net on RLED instead of ESIM (ablation) | NER-Net · DSEC-night | LOE: 1421.2 -> 1031.2 (LOE) | Improved |
| replace E2VID+ with NER-Net (both RLED-trained) | E2VID+ · MVSEC-night | LOE: 1283.8 -> 1198.4 (LOE) | Improved |
| replace SCI with NER-Net on nighttime dynamic scenes | SCI · Nighttime dynamic scenes | LOE: 99.78 -> 92.81 (LOE) | Improved |
| replace URetinex-Net with NER-Net on SPAQ | URetinex-Net · Nighttime dynamic scenes | SPAQ: 9228.63 -> 8205.78 | Hurt |
Closest papers by meaning
- Seeing Motion at Nighttime with an Event Camera (CVPR 2024 · similarity 0.98)
- Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approach (CVPR 2024 · similarity 0.88)
- AE-NeRF: Augmenting Event-Based Neural Radiance Fields for Non-ideal Conditions and Larger Scenes (AAAI 2025 · similarity 0.86)
- LED: A Large-scale Real-world Paired Dataset for Event Camera Denoising (CVPR 2024 · similarity 0.86)
- Event-3DGS: Event-based 3D Reconstruction Using 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.86)
- Mitigating Motion Blur in Neural Radiance Fields with Events and Frames (CVPR 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.
