Research map / Video / Video generation
InverseCrafter: Efficient Video ReCapture as a Latent Domain Inverse Problem
Yeobin Hong, Suhyeon Lee, 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 |
|---|---|---|---|
| latent continuous mask + DDS guidance vs fine-tuned/optimization baselines | InverseCrafter · UltraVideo camera control | PSNR: 28.37 -> 29.35 | Improved |
| latent mask guidance instead of fine-tuned model | InverseCrafter · UltraVideo camera control | runtime: 696 s (NVS-Solver) -> 71 s | Improved |
| add measurement guidance to base VDM (ablation) | InverseCrafter · UltraVideo | runtime: 70 -> 71 s | No clear effect |
| training-free proxy mask instead of learned mask encoder (ablation) | InverseCrafter · DAVIS novel trajectories | PSNR: 25.60 -> 25.48 | No clear effect |
| swap DDS for FlowDPS solver (ablation) | InverseCrafter · DAVIS novel trajectories | PSNR: 25.60 -> 27.43 | Improved |
| swap backbone to CogVideoX (ablation) | InverseCrafter · DAVIS novel trajectories | PSNR: 25.60 -> 26.46 | Improved |
| apply to text-guided inpainting vs VideoPainter | InverseCrafter · DAVIS text-guided inpainting | FVD: 1432.77 -> 1422.72 | Improved |
| apply to text-guided inpainting vs Zero4D | InverseCrafter · DAVIS text-guided inpainting | CLIP score: 25.83 vs 24.98 | Hurt |
| continuous C-channel latent mask instead of binary resized mask (ablation) | InverseCrafter · DAVIS subset | measurement consistency | Improved |
| larger alpha (earlier measurement consistency step) (ablation) | InverseCrafter · UltraVideo | consistency vs fidelity | Mixed |
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 3 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
- Dynamic View Synthesis as an Inverse Problem (NeurIPS 2025 · similarity 0.89)
- Training-Free Efficient Video Generation via Dynamic Token Carving (NeurIPS 2025 · similarity 0.87)
- Reconstruct, Inpaint, Test-Time Finetune: Dynamic Novel-view Synthesis from Monocular Videos (NeurIPS 2025 · similarity 0.87)
- Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models (CVPR 2025 · similarity 0.86)
- NVS-Solver: Video Diffusion Model as Zero-Shot Novel View Synthesizer (ICLR 2025 · similarity 0.86)
- GenFusion: Closing the Loop between Reconstruction and Generation via Videos (CVPR 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.
