Research map / Video / Video generation
3D Scene-Adaptive Trajectory-Controllable Human Image Animation with Camera Movement
Deyin Liu, Jicheng Xu, Lin Yuanbo Wu, Xiaowei Zhao, Xiatian Zhu, Anjan Dutta, Zhe Jin
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 ground-adaptive 3D motion retargeting (ablation) | 3STC-HIA · RealisDance-Val | FID: 41.65 -> 32.78 | Improved |
| add ground-adaptive 3D motion retargeting (ablation) | 3STC-HIA · RealisDance-Val | FVD: 685.42 -> 521.37 | Improved |
| add ground-adaptive 3D motion retargeting (ablation) | 3STC-HIA · RealisDance-Val | Motion smoothness: 93.28 -> 97.42 | Improved |
| add ground-adaptive 3D motion retargeting (ablation) | 3STC-HIA · RealisDance-Val | Dynamic degree: 65.73 -> 68.95 | Improved |
| add viewpoint-adaptive scene-guided fusion (ablation) | 3STC-HIA · RealisDance-Val | Background consistency: 90.23 -> 95.64 | Improved |
| add viewpoint-adaptive scene-guided fusion (ablation) | 3STC-HIA · RealisDance-Val | FID: 58.47 -> 32.78 | Improved |
| add viewpoint-adaptive scene-guided fusion (ablation) | 3STC-HIA · RealisDance-Val | FVD: 832.15 -> 521.37 | Improved |
| add viewpoint-adaptive scene-guided fusion (ablation) | 3STC-HIA · RealisDance-Val | Temporal flicker: 90.56 -> 96.89 | Improved |
| add viewpoint-adaptive scene-guided fusion (ablation) | 3STC-HIA · RealisDance-Val | Aesthetic quality: 53.28 -> 57.86 | Improved |
| add ground-adaptive trajectory control and scene fusion vs RealisMotion | 3STC-HIA · Trajectory100 | Translation error: 1.305 -> 0.478 | Improved |
| 3STC-HIA vs RealisMotion rotation error | 3STC-HIA · Trajectory100 | Rotation error: 0.095 -> 0.162 | Hurt |
| 3STC-HIA vs RealisDance FVD | 3STC-HIA · RealisDance-Val | FVD: 305.84 -> 573.41 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- RealisMotion — RealisMotion: Decomposed Human Motion Control and Video Generation in the World Space (ICML 2026) · better on 3 of 6 shared results
- MimicMotion — MimicMotion: High-Quality Human Motion Video Generation with Confidence-aware Pose Guidance (ICML 2025) · better on 2 of 2 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
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.88)
- DisPose: Disentangling Pose Guidance for Controllable Human Image Animation (ICLR 2025 · similarity 0.88)
- 3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation (ICLR 2025 · similarity 0.88)
- Move-in-2D: 2D-Conditioned Human Motion Generation (CVPR 2025 · similarity 0.88)
- Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance (NeurIPS 2025 · similarity 0.87)
- Sitcom-Crafter: A Plot-Driven Human Motion Generation System in 3D Scenes (ICLR 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.
