Research map / 3D vision / Depth and stereo
Unsupervised Joint Learning of Optical Flow and Intensity with Event Cameras
Shuang Guo, Friedhelm Hamann, Guillermo Gallego
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 temporal consistency loss (ablation) | joint flow-intensity network · DSEC | EPE: 2.006 -> 1.781 | Improved |
| add temporal consistency loss (ablation) | joint flow-intensity network · BS-ERGB | LPIPS: 0.637 -> 0.555 | Improved |
| add temporal consistency loss (ablation) | joint flow-intensity network · BS-ERGB | MSE: 0.160 -> 0.100 | Improved |
| add TV regularization of flow and intensity (ablation) | joint flow-intensity network · DSEC | EPE: 3.303 -> 1.781 | Improved |
| add TV regularization of flow and intensity (ablation) | joint flow-intensity network · BS-ERGB | MSE: 0.098 -> 0.100 | No clear effect |
| add joint intensity estimation to flow training (ablation) | CMax-based flow network · DSEC | EPE: 3.134 -> 1.781 | Improved |
| use a single network instead of two separate networks (ablation) | joint estimation · DSEC | EPE: 3.30 -> 1.781 | Improved |
| use a single network instead of two separate networks (ablation) | joint estimation · BS-ERGB | LPIPS: 0.59 -> 0.555 | Improved |
| decrease CMax weight lambda2 to 0.2 (ablation) | joint network · DSEC | EPE: 1.781 -> 2.777 | Hurt |
| increase CMax weight lambda2 to 5.0 (ablation) | joint network · DSEC | EPE: 1.781 -> 2.34 | Hurt |
| change contrast threshold C from 0.2 to 0.1 (ablation) | joint network · DSEC | EPE: 1.781 -> 1.893 | No clear effect |
| change contrast threshold C from 0.2 to 0.4 (ablation) | joint network · DSEC | EPE: 1.781 -> 1.884 | No clear effect |
Closest papers by meaning
- Unsupervised 3d Motion Estimation Using Event Camera (CVPR 2026 · similarity 0.91)
- Simultaneous Optical Flow and Intensity Estimation From an Event Camera (CVPR 2016 · similarity 0.90)
- Secrets of Event-Based Optical Flow (ECCV 2022 · similarity 0.90)
- Single Image Optical Flow Estimation With an Event Camera (CVPR 2020 · similarity 0.90)
- Injecting Frame-Event Complementary Fusion into Diffusion for Optical Flow in Challenging Scenes (NeurIPS 2025 · similarity 0.88)
- EDCFlow: Exploring Temporally Dense Difference Maps for Event-based Optical Flow Estimation (CVPR 2025 · similarity 0.88)
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.
