Research map / Video / Video segmentation and tracking
Event Stream Filtering via Probability Flux Estimation
Jinze Chen, Wei Zhai, Yang Cao, Bin Li, Zheng-Jun Zha
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 |
|---|---|---|---|
| use EDFilter instead of raw events for denoising | EDFilter · RED circle1 (EVK4) | NMSE: 0.390 -> 0.193 | Improved |
| use frame-binned EventZoom instead of continuous-time filtering | EventZoom · RED NMSE | NMSE: 0.331 vs 0.121 (spiral1 DAVIS346, EDFilter) | Hurt |
| aggressive motion-based filtering (EvFlow) for tracking | EvFlow · RED point tracking | ATE (mm): 2.290 (DAVIS346 point); NMSE 0.285 on circle1 DAVIS346 | Mixed |
| replace EDF temporal model with Poisson or Gaussian temporal models (ablation) | EDFilter · RED filtering and point tracking | NMSE and ATE | Hurt |
| use L1 instead of L2 spatial reconstruction (ablation) | EDFilter · RED | ATE vs NMSE | Mixed |
| asynchronous sampling instead of hybrid 10ms sampling (ablation) | EDFilter · RED | ATE vs NMSE | Mixed |
| synchronous sampling instead of hybrid sampling (ablation) | EDFilter · RED | NMSE and ATE | Hurt |
| finer synchronous sampling (1ms vs 10ms) (ablation) | EDFilter · RED | NMSE and ATE | No clear effect |
| insert EDFilter before ESVO SLAM | ESVO · MVSEC and RPG | RMS error (cm) | Improved |
| fine-tune E2VID on EDFilter-filtered events | E2VID · E-MLB | PSNR, MSE, SSIM, LPIPS | Improved |
| use EDFilter instead of EventZoom for 2x super-resolution | EDFilter · RED 2x super-resolution | NMSE: 0.190 -> 0.103 (circle1 DAVIS346) | Improved |
Closest papers by meaning
- EKLT: Asynchronous Photometric Feature Tracking Using Events and Frames (IJCV 2019 · similarity 0.86)
- Event-Illumination Collaborative Low-light Image Enhancement with a High-resolution Real-world Dataset (CVPR 2026 · similarity 0.86)
- Seeing Motion at Nighttime with an Event Camera (CVPR 2024 · similarity 0.86)
- Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approach (CVPR 2024 · similarity 0.85)
- High Speed and High Dynamic Range Video with an Event Camera (TPAMI 2019 · similarity 0.85)
- LED: A Large-scale Real-world Paired Dataset for Event Camera Denoising (CVPR 2024 · 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.
