Research map / Video / Video segmentation and tracking
Edge-Guided Fusion and Motion Augmentation for Event-Image Stereo
Fengan Zhao, Qianang Zhou, Junlin Xiong
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 edge-guided event-image fusion (EGEIF) instead of concatenation (ablation) | RAFT-based event-image stereo baseline · MVSEC split 1 | mean disparity error (pix): 0.3936 -> 0.3616 | Improved |
| add EGEIF (ablation) | RAFT-based baseline · MVSEC split 3 | mean disparity error (pix): 0.4696 -> 0.4013 | Improved |
| add edge-guided motion augmentation (EGMA) update module (ablation) | RAFT-based baseline · MVSEC split 1 | mean disparity error (pix): 0.3936 -> 0.3728 | Improved |
| add EGMA alone (ablation) | RAFT-based baseline · MVSEC split 3 | mean disparity error (pix): 0.4696 -> 0.4641 | No clear effect |
| combine EGEIF and EGMA (ablation) | RAFT-based baseline · MVSEC split 3 | mean disparity error (pix): 0.4013 -> 0.3818 | Improved |
| combine EGEIF and EGMA (ablation) | RAFT-based baseline · MVSEC split 1 | mean depth error (cm): 11.3504 -> 9.7240 | Improved |
| edge-guided method versus previous best event-image method SCS-Net | Event-image stereo · MVSEC split 3 | one-pixel accuracy (%): 94.0 -> 93.4 | Hurt |
| edge-guided method versus SCS-Net | Event-image stereo · MVSEC split 3 | mean depth error (cm): 13.5 -> 11.1 | Improved |
Closest papers by meaning
- Event-Intensity Stereo: Estimating Depth by the Best of Both Worlds (ICCV 2021 · similarity 0.91)
- Event-Image Fusion Stereo Using Cross-Modality Feature Propagation (AAAI 2022 · similarity 0.91)
- Deep Event Stereo Leveraged by Event-to-Image Translation (AAAI 2021 · similarity 0.91)
- Zero-Shot Event-Intensity Asymmetric Stereo via Visual Prompting from Image Domain (NeurIPS 2024 · similarity 0.88)
- Active Event-based Stereo Vision (CVPR 2025 · similarity 0.87)
- DERD-Net: Learning Depth from Event-based Ray Densities (NeurIPS 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.
