Research map / 3D vision / Depth and stereo
x^2-Fusion: Cross-Modality and Cross-Dimension Flow Estimation in Event Edge Space
Ruishan Guo, Ciyu Ruan, Haoyang Wang, Zihang Gong, Jingao Xu, Xinlei Chen
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 Event Edge Space (ablation) | x^2-Fusion · DSEC val | EPE 2D: 0.491 -> 0.322 | Improved |
| add Event Edge Space (ablation) | x^2-Fusion · DSEC val | EPE 3D: 0.146 -> 0.094 | Improved |
| add edge-anchored symmetric regularization (ablation) | x^2-Fusion · DSEC val | EPE 2D: 0.393 -> 0.322 | Improved |
| add edge-anchored symmetric regularization (ablation) | x^2-Fusion · DSEC val | EPE 3D: 0.119 -> 0.094 | Improved |
| pretrain the event edge encoder (ablation) | x^2-Fusion · DSEC val | EPE 2D: 0.378 -> 0.322 | Improved |
| pretrain the event edge encoder (ablation) | x^2-Fusion · DSEC val | EPE 3D: 0.114 -> 0.094 | Improved |
| joint 2D and 3D estimation instead of independent (ablation) | x^2-Fusion · DSEC val | EPE 2D: 0.404 -> 0.386 | Improved |
| joint 2D and 3D estimation instead of independent (ablation) | x^2-Fusion · DSEC val | EPE 3D: 0.119 -> 0.113 | Improved |
| add cross-dimension contrastive learning (CCL) to joint training (ablation) | x^2-Fusion · DSEC val | EPE 2D: 0.386 -> 0.325 | Improved |
| add cross-dimension contrastive learning (CCL) to joint training (ablation) | x^2-Fusion · DSEC val | EPE 3D: 0.113 -> 0.103 | Improved |
| use x^2-Fusion instead of RPEFlow under drifting LiDAR | x^2-Fusion · DSEC-PC drifting | EPE 2D: 1.051 -> 0.409 | Improved |
| use x^2-Fusion instead of RPEFlow under under-exposure | x^2-Fusion · EKubric-Img under-exposure | EPE 2D: 3.663 -> 1.143 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- RPEFlow — RPEFlow: Multimodal Fusion of RGB-PointCloud-Event for Joint Optical Flow and Scene Flow Estimation (ICCV 2023) · better on 9 of 9 shared results
- CamLiFlow — CamLiFlow: Bidirectional Camera-LiDAR Fusion for Joint Optical Flow and Scene Flow Estimation (CVPR 2022) · better on 4 of 4 shared results
- RAFT-3D — RAFT-3D: Scene Flow Using Rigid-Motion Embeddings (CVPR 2021) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Bring Event into RGB and LiDAR: Hierarchical Visual-Motion Fusion for Scene Flow (CVPR 2024 · similarity 0.90)
- Bridge Frame and Event: Common Spatiotemporal Fusion for High-Dynamic Scene Optical Flow (CVPR 2025 · similarity 0.88)
- Cross-Modal Distillation for 2D/3D Multi-Object Discovery from 2D Motion (CVPR 2025 · similarity 0.87)
- Injecting Frame-Event Complementary Fusion into Diffusion for Optical Flow in Challenging Scenes (NeurIPS 2025 · similarity 0.87)
- X-Drive: Cross-modality Consistent Multi-Sensor Data Synthesis for Driving Scenarios (ICLR 2025 · similarity 0.86)
- Event-Driven Dynamic Scene Depth Completion (NeurIPS 2025 · 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.
