Research map / 3D vision / 3D generation
Video Camera Trajectory Editing with Generative Rendering from Estimated Geometry
Junyoung Seo, Jisang Han, Jaewoo Jung, Siyoon Jin, JoungBin Lee, Takuya Narihira, Kazumi Fukuda, Takashi Shibuya, Donghoon Ahn, Shoukang Hu, Seungryong Kim, Yuki Mitsufuji
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 geometry grounding via 2D flow instead of camera-pose injection (ablation) | Vid-CamEdit · Neu3D | LPIPS: 0.498 -> 0.414 | Improved |
| add geometry grounding via 2D flow instead of camera-pose injection (ablation) | Vid-CamEdit · Neu3D | SSIM: 0.314 -> 0.358 | Improved |
| use generative rendering instead of reprojection plus inpainting (ablation) | Vid-CamEdit · Neu3D | LPIPS: 0.485 -> 0.414 | Improved |
| swap MonST3R for DepthCrafter geometry (ablation) | Vid-CamEdit · Neu3D | LPIPS: 0.414 -> 0.416 | No clear effect |
| swap MonST3R for DepthAnyVideo geometry (ablation) | Vid-CamEdit · Neu3D | LPIPS: 0.414 -> 0.425 | No clear effect |
| use Vid-CamEdit instead of Generative Camera Dolly (ablation) | Vid-CamEdit · ST-NeRF | Frame consistency (CLIP): 0.748 -> 0.917 | Improved |
| use Vid-CamEdit instead of Pseudo-DVS | Vid-CamEdit · ST-NeRF | PSNR: 15.33 -> 14.89 | Hurt |
| add Vid-CamEdit supervision to Shape-of-Motion | Shape-of-Motion · DyCheck (occluded) | LPIPS: 0.465 -> 0.129 | Improved |
| add Vid-CamEdit supervision to Shape-of-Motion | Shape-of-Motion · DyCheck (co-visible) | SSIM: 0.646 -> 0.573 | Hurt |
Closest papers by meaning
- LaVR: Scene Latent Conditioned Generative Video Trajectory Re-Rendering using Large 4D Reconstruction Models (CVPR 2026 · similarity 0.90)
- UniGeo: Unifying Geometric Constraints for Camera-Controllable Image Editing via Video Priors (ECCV 2026 · similarity 0.88)
- Geometry-Aware Single-Image 4D Synthesis via Dense Trajectory Generation (ECCV 2026 · similarity 0.88)
- Learning to Correct Geometry in Generated Videos (NeurIPS 2026 · similarity 0.88)
- World from Motion: Generative Dynamic Gaussian Reconstruction from Monocular Video (NeurIPS 2026 · similarity 0.88)
- SimVS: Simulating World Inconsistencies for Robust View Synthesis (CVPR 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.
