Research map / Video / Video generation
Causal-Entity Reflected Egocentric Traffic Accident Video Synthesis
Lei-Lei Li, Jianwu Fang, Junbin Xiao, Shanmin Pang, Hongkai Yu, Chen Lv, Jianru Xue, Tat-Seng Chua
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 driver gaze maps (ablation) | Causal-VidSyn · N2A (BDD-A, 2000 clips) | CLIPs: 26.0 -> 26.5 | Improved |
| add driver gaze maps (ablation) | Causal-VidSyn · N2A (BDD-A, 2000 clips) | FVD: 6249.3 -> 6192.3 | Improved |
| add driver gaze maps (ablation) | Causal-VidSyn · AEdit (DADA-2000, 3000 clips) | Afd (%): 50.4 -> 55.4 | Improved |
| add CTS and CTG (ablation) | Causal-VidSyn · N2A (BDD-A, 2000 clips) | CLIPs: 25.0 -> 26.0 | Improved |
| add CTS and CTG (ablation) | Causal-VidSyn · N2A (BDD-A, 2000 clips) | FVD: 7225.5 -> 6249.3 | Improved |
| add CTS and CTG (ablation) | Causal-VidSyn · AEdit (DADA-2000, 3000 clips) | CLIPs: 28.6 -> 28.6 | No clear effect |
| add RPFD (ablation) | Causal-VidSyn · AEdit (DADA-2000, 3000 clips) | CLIPs: 26.9 -> 28.6 | Improved |
| add RPFD (ablation) | Causal-VidSyn · N2A (BDD-A, 2000 clips) | FVD: 6378.9 -> 7225.5 | Hurt |
| add CTS to downscale layers (ablation) | Causal-VidSyn · AEdit (DADA-2000, 3000 clips) | Afd (%): 49.9 -> 55.4 | Improved |
| add CTS to upscale layers (ablation) | Causal-VidSyn · AEdit (DADA-2000, 3000 clips) | Afd (%): 54.9 -> 55.4 | Improved |
| add inner CTS blocks to CTG-only model (ablation) | Causal-VidSyn · AEdit (DADA-2000, 3000 clips) | Afd (%): 50.2 -> 55.4 | Improved |
| add CTS and CTG to fine-tune CogV-X-T | CogV-X-T · AEdit (DADA-2000, 3000 clips) | CLIPs: +2.9 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- LAMP — LAMP: Learn A Motion Pattern for Few-Shot Video Generation (CVPR 2024) · better on 2 of 3 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
- Abductive Ego-View Accident Video Understanding for Safe Driving Perception (CVPR 2024 · similarity 0.89)
- Traffic Accident Benchmark for Causality Recognition (ECCV 2020 · similarity 0.88)
- Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving (NeurIPS 2025 · similarity 0.84)
- Robust Egocentric Referring Video Object Segmentation via Dual-Modal Causal Intervention (NeurIPS 2025 · similarity 0.84)
- iFinder: Structured Zero-Shot Vision-Based LLM Grounding for Dash-Cam Video Reasoning (NeurIPS 2025 · similarity 0.83)
- Transfer Your Perspective: Controllable 3D Generation from Any Viewpoint in a Driving Scene (CVPR 2025 · similarity 0.83)
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.
