Research map / Video / Video generation
FreeTraj: Tuning-Free Trajectory Control via Noise Guided Video Diffusion
Haonan Qiu, Zhaoxi Chen, Zhouxia Wang, Yingqing He, Menghan Xia, Ziwei Liu
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 noise guidance to attention guidance (ablation) | FreeTraj · User study | trajectory alignment votes: 19.75% -> 65.74% | Improved |
| add noise guidance to attention guidance (ablation) | FreeTraj · User study | video-text alignment votes: 15.43% -> 62.65% | Improved |
| add noise guidance to attention guidance (ablation) | FreeTraj · User study | video quality votes: 15.74% -> 63.89% | Improved |
| add noise guidance to attention guidance (ablation) | FreeTraj · Qualitative trajectory control | trajectory accuracy | Improved |
| use soft attention guidance instead of hard attention mask (ablation) | FreeTraj · Qualitative video quality | artifacts | Improved |
| add high-frequency noise resampling in trajectory injection (ablation) | FreeTraj · Qualitative video quality | artifacts | Improved |
| FreeTraj versus Peekaboo | FreeTraj · Trajectory control | mIoU: 0.235 -> 0.281 | Improved |
| FreeTraj versus Peekaboo | FreeTraj · Trajectory control | CD: 0.179 -> 0.154 | Improved |
| FreeTraj versus Peekaboo | FreeTraj · Trajectory control | FVD: 403.00 -> 436.22 | Hurt |
| FreeTraj versus Direct inference | FreeTraj · Trajectory control | mIoU: 0.161 -> 0.281 | Improved |
Later papers that compared with this method
3 later accepted papers report FreeTraj* as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- ObjCtrl-2.5D: Training-free Object Control with Camera Poses (IJCV 2026 · better on 6 of 6 shared results)
- Zo3T: Zero-Shot 3D-Aware Trajectory-Guided Image-to-Video Generation via Test-Time Training (AAAI 2026 · better on 1 of 1 shared result)
- QWERTY: Training-Free Motion Control via Query-Warped Video Diffusion Transformers (ECCV 2026 · better on 1 of 1 shared result)
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
- Motion-Zero: A Zero-Shot Trajectory Control Framework of Moving Object for Diffusion-Based Video Generation (AAAI 2025 · similarity 0.90)
- Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise (CVPR 2025 · similarity 0.90)
- FreeNoise: Tuning-Free Longer Video Diffusion via Noise Rescheduling (ICLR 2024 · similarity 0.89)
- Training-free Camera Control for Video Generation (ICLR 2025 · similarity 0.88)
- Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance (NeurIPS 2025 · similarity 0.88)
- Trajectory attention for fine-grained video motion control (ICLR 2025 · similarity 0.88)
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.
