Research map / Video / Video generation
GimbalDiffusion: Gravity-Aware Camera Control for Video Generation
Frédéric Fortier-Chouinard, Yannick Hold-Geoffroy, Valentin Deschaintre, Matheus Gadelha, Jean-Francois Lalonde
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 null-pitch conditioning (caption upright crops) (ablation) | GimbalDiffusion WAN 2.2 5B · SpatialVID-extreme | PitchErr: 14.39 -> 8.36 | Improved |
| add null-pitch conditioning (ablation) | GimbalDiffusion WAN 2.1 1.3B UCPE · SpatialVID-extreme | PitchErr: 19.13 -> 13.95 | Improved |
| add null-pitch conditioning (ablation) | GimbalDiffusion WAN 2.2 5B · SpatialVID-extreme | RotErr: 17.26 -> 15.05 | Improved |
| add null-pitch conditioning (ablation) | GimbalDiffusion WAN 2.1 1.3B UCPE · SpatialVID-extreme | RotErr: 11.68 -> 12.26 | Hurt |
| add null-pitch conditioning (ablation) | GimbalDiffusion WAN 2.2 5B · SpatialVID-extreme | CLIP: 22.0 -> 20.6 | Hurt |
| use absolute rather than relative camera rotations (ablation) | GimbalDiffusion UCPE · SpatialVID-extreme | PitchErr: 51.16 -> 13.95 | Improved |
| use absolute gravity-aligned conditioning on 360 video data instead of UCPE | UCPE · SpatialVID-extreme | PitchErr: 16.25 -> 13.95 | Improved |
| use absolute gravity conditioning instead of AC3D with camera text | AC3D · SpatialVID-extreme | PitchErr: 39.09 -> 19.67 | Improved |
| use absolute gravity conditioning instead of PreciseCam+WAN | PreciseCam+WAN-I2V-CC · SpatialVID-extreme | PitchErr: 29.16 -> 10.50 | Improved |
| use GimbalDiffusion instead of PreciseCam+WAN on visual quality (ablation) | PreciseCam+WAN-I2V-CC · SpatialVID-extreme | FVD: 776.3 -> 1041.6 | Hurt |
| use GimbalDiffusion instead of PreciseCam+WAN on relative rotation (ablation) | PreciseCam+WAN-I2V-CC · SpatialVID-extreme | RotErr: 24.54 -> 25.84 | Hurt |
| add null-pitch conditioning (ablation) | GimbalDiffusion · PolyHaven prompt-camera entanglement | pitch error / CLIP to GT | Improved |
Closest papers by meaning
- Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance (NeurIPS 2025 · similarity 0.88)
- AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers (CVPR 2025 · similarity 0.87)
- Trajectory attention for fine-grained video motion control (ICLR 2025 · similarity 0.87)
- SG-I2V: Self-Guided Trajectory Control in Image-to-Video Generation (ICLR 2025 · similarity 0.87)
- Motion Prompting: Controlling Video Generation with Motion Trajectories (CVPR 2025 · similarity 0.87)
- MotionPro: A Precise Motion Controller for Image-to-Video Generation (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.
