Research map / Video / Video generation
Zo3T: Zero-Shot 3D-Aware Trajectory-Guided Image-to-Video Generation via Test-Time Training
Ruicheng Zhang, Jun Zhou, Zunnan Xu, Zihao Liu, Jiehui Huang, Mingyang Zhang, Yu Sun, Xiu 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 3D-aware kinematic projection (perspective-scaled boxes) (ablation) | Zo3T · VIPSeg | FID: 76.12 -> 74.83 | Improved |
| add 3D-aware kinematic projection (ablation) | Zo3T · VIPSeg | ObjMC: 12.98 -> 12.74 | Improved |
| add test-time LoRA adaptation instead of hard latent editing (ablation) | Zo3T · VIPSeg | FID: 89.36 -> 74.83 | Improved |
| add test-time LoRA adaptation instead of hard latent editing (ablation) | Zo3T · VIPSeg | ObjMC: 14.24 -> 12.74 | Improved |
| add Guidance Field Rectification (ablation) | Zo3T · VIPSeg | FVD: 205.18 -> 197.63 | Improved |
| add Guidance Field Rectification (ablation) | Zo3T · VIPSeg | ObjMC: 13.92 -> 12.74 | Improved |
| add Fourier Orthogonal Recomposition (ablation) | Zo3T · VIPSeg | FID: 95.45 -> 74.83 | Improved |
| add Fourier Orthogonal Recomposition (ablation) | Zo3T · VIPSeg | ObjMC: 12.81 -> 12.74 | No clear effect |
| zero-shot Zo3T versus supervised trajectory methods on motion precision | Zo3T · VIPSeg | ObjMC: 10.84 (DragNUWA) vs 12.74 | Hurt |
| zero-shot Zo3T versus SG-I2V | Zo3T · VIPSeg | ObjMC: 14.43 -> 12.74 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- ObjCtrl-2.5D — ObjCtrl-2.5D: Training-free Object Control with Camera Poses (IJCV 2026) · better on 5 of 5 shared results
- LeviTor — LeviTor: 3D Trajectory Oriented Image-to-Video Synthesis (CVPR 2025) · better on 3 of 4 shared results
- SG-I2V — SG-I2V: Self-Guided Trajectory Control in Image-to-Video Generation (ICLR 2025) · better on 3 of 3 shared results
- FreeTraj — FreeTraj: Tuning-Free Trajectory Control via Noise Guided Video Diffusion (IJCV 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
- SG-I2V: Self-Guided Trajectory Control in Image-to-Video Generation (ICLR 2025 · similarity 0.88)
- Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance (NeurIPS 2025 · similarity 0.88)
- PoseTraj: Pose-Aware Trajectory Control in Video Diffusion (CVPR 2025 · similarity 0.88)
- Motion-Zero: A Zero-Shot Trajectory Control Framework of Moving Object for Diffusion-Based Video Generation (AAAI 2025 · similarity 0.87)
- 3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation (ICLR 2025 · similarity 0.87)
- GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control (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.
