Research map / Video / Video generation
RealCam-I2V: Real-World Image-to-Video Generation with Interactive Complex Camera Control
Teng Li, Guangcong Zheng, Rui Jiang, Shuigen Zhan, Tao Wu, Yehao Lu, Yining Lin, Chuanyun Deng, Yepan Xiong, Min Chen, Lin Cheng, Xi Li
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 metric scene-scale alignment (MSA) to CamI2V (ablation) | CamI2V* · RealEstate10K test (1000 clips) | TransErr (metric scale): 3.2934 -> 2.6524 | Improved |
| add MSA to MotionCtrl (ablation) | MotionCtrl* · RealEstate10K test (1000 clips) | TransErr (metric scale): 6.8182 -> 4.2218 | Improved |
| add MSA to CameraCtrl (ablation) | CameraCtrl* · RealEstate10K test (1000 clips) | TransErr (metric scale): 5.5090 -> 3.8218 | Improved |
| add MSA to CamI2V (ablation) | CamI2V* · RealEstate10K test (1000 clips) | RotErr: 0.4120 -> 0.4243 | No clear effect |
| add scene-constrained noise shaping (SNS) to DynamiCrafter (ablation) | DynamiCrafter · RealEstate10K test (1000 clips) | RotErr: 3.3415 -> 1.5163 | Improved |
| add SNS to MSA-trained MotionCtrl (ablation) | MotionCtrl* · RealEstate10K test (1000 clips) | FVD (VideoGPT): 67.130 -> 58.885 | Improved |
| add SNS to MSA-trained CameraCtrl (ablation) | CameraCtrl* · RealEstate10K test (1000 clips) | TransErr (metric scale): 3.8218 -> 3.1845 | Improved |
| add SNS to MSA-trained CamI2V (full RealCam-I2V) (ablation) | CamI2V* · RealEstate10K test (1000 clips) | TransErr (metric scale): 2.6524 -> 2.2317 | Improved |
| add SNS to RealCam-I2V (ablation) | RealCam-I2V · VBench-I2V (RealEstate10K-trained, DynamiCrafter) | Camera Motion: 87.42 -> 93.32 | Improved |
| add SNS to RealCam-I2V (ablation) | RealCam-I2V · VBench-I2V (RealEstate10K-trained, DynamiCrafter) | Dynamic Degree: 46.75 -> 35.77 | Hurt |
| add MSA to RealCam-I2V (ablation) | RealCam-I2V · VBench-I2V (RealEstate10K-trained, DynamiCrafter) | Total Score: 83.77 -> 85.71 | Improved |
| lower noise shaping threshold from 900 to 600 (apply SNS longer) (ablation) | RealCam-I2V · VBench-I2V (RealEstate10K-trained, DynamiCrafter) | Dynamic Degree: 35.77 -> 27.64 | Hurt |
Later papers that compared with this method
1 later accepted paper reports Real-CamI2V as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Captain Safari: A World Engine with Pose-Aligned 3D Memory (CVPR 2026 · better on 8 of 8 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.
- DynamiCrafter — DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors (ECCV 2024) · better on 8 of 8 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
- AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers (CVPR 2025 · similarity 0.88)
- Director3D: Real-world Camera Trajectory and 3D Scene Generation from Text (NeurIPS 2024 · similarity 0.87)
- Training-free Camera Control for Video Generation (ICLR 2025 · similarity 0.87)
- GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control (CVPR 2025 · similarity 0.87)
- Extrapolating and Decoupling Image-to-Video Generation Models: Motion Modeling is Easier Than You Think (CVPR 2025 · similarity 0.87)
- 4Real: Towards Photorealistic 4D Scene Generation via Video Diffusion Models (NeurIPS 2024 · 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.
