Research map / Video / Video generation
ReDirector: Creating Any-Length Video Retakes with Rotary Camera Encoding
Byeongjun Park, Byung-Hoon Kim, Hyungjin Chung, Jong Chul Ye
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 |
|---|---|---|---|
| share 3D RoPE and use Plucker ray token-wise camera encoding (vs ReCamMaster) (ablation) | ReDirector with additive conditioning · DAVIS retakes (50 videos x 10 trajectories) | Dyn-MEt3R: 0.7857 -> 0.8378 | Improved |
| share 3D RoPE and Plucker rays (vs ReCamMaster) (ablation) | ReDirector with additive conditioning · DAVIS retakes (50 videos x 10 trajectories) | RotErr: 2.347 -> 1.975 | Improved |
| replace additive conditioning with RoCE (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | RotErr: 1.975 -> 1.897 | Improved |
| replace additive conditioning with RoCE (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | Dyn-MEt3R: 0.8378 -> 0.8341 | Hurt |
| replace additive conditioning with RoCE (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | Aesthetic Quality: 0.5121 -> 0.5139 | Improved |
| add geometry-aware attention (GTA) on top of RoCE (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | Dyn-MEt3R: 0.8341 -> 0.8477 | Improved |
| add geometry-aware attention (GTA) on top of RoCE (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | MEt3R: 0.3164 -> 0.3073 | Improved |
| add geometry-aware attention (GTA) on top of RoCE (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | RotErr: 1.897 -> 1.666 | Improved |
| use ViPE metric-scale poses instead of MegaSaM (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | Dyn-MEt3R: 0.8357 -> 0.8477 | Improved |
| use ViPE metric-scale poses instead of MegaSaM (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | RotErr: 1.942 -> 1.666 | Improved |
| train 50K iterations instead of 20K (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | RotErr: 1.666 -> 1.521 | Improved |
| train 50K iterations instead of 20K (ablation) | ReDirector · DAVIS retakes (50 videos x 10 trajectories) | MEt3R: 0.3073 -> 0.2954 | Improved |
Later papers that compared with this method
1 later accepted paper reports ReDirector as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SierpinskiCam: Camera-Controlled Video Retaking with Sierpinski Triangle Pattern Cues (NeurIPS 2026 · better on 8 of 12 shared results)
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.
- ReCamMaster — ReCamMaster: Camera-Controlled Generative Rendering from A Single Video (ICCV 2025) · better on 7 of 7 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
- SierpinskiCam: Camera-Controlled Video Retaking with Sierpinski Triangle Pattern Cues (NeurIPS 2026 · similarity 0.85)
- VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control (ICLR 2025 · similarity 0.83)
- AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers (CVPR 2025 · similarity 0.83)
- Trajectory attention for fine-grained video motion control (ICLR 2025 · similarity 0.83)
- Cinematic Behavior Transfer via NeRF-based Differentiable Filming (CVPR 2024 · similarity 0.83)
- Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance (NeurIPS 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.
