Research map / Video / Video generation
DIMO: Diverse 3D Motion Generation for Arbitrary Objects
Linzhan Mou, Jiahui Lei, Chen Wang, Lingjie Liu, Kostas Daniilidis
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 neural keypoint motion factorization (ablation) | DIMO · Ablation (Consistent4D-style metrics) | FVD: 851.83 -> 587.09 | Improved |
| add neural keypoint motion factorization (ablation) | DIMO · Ablation (Consistent4D-style metrics) | LPIPS: 0.134 -> 0.126 | Improved |
| add motion latent space (ablation) | DIMO · Ablation (Consistent4D-style metrics) | FVD: 1077.19 -> 587.09 | Improved |
| add motion latent space (ablation) | DIMO · Ablation (Consistent4D-style metrics) | LPIPS: 0.163 -> 0.126 | Improved |
| add motion pre-training (ablation) | DIMO · Ablation (Consistent4D-style metrics) | FVD: 890.26 -> 587.09 | Improved |
| add motion pre-training (ablation) | DIMO · Ablation (Consistent4D-style metrics) | CLIP: 0.90 -> 0.93 | Improved |
| add multi-motion joint training (ablation) | DIMO · Ablation (Consistent4D-style metrics) | FVD: 693.49 -> 587.09 | Improved |
| add multi-motion joint training (ablation) | DIMO · Ablation (Consistent4D-style metrics) | LPIPS: 0.131 -> 0.126 | Improved |
| use DIMO instead of SV4D | SV4D · Video-to-4D (Consistent4D benchmark) | FVD: 732.40 -> 625.30 | Improved |
| use DIMO instead of SV4D | SV4D · Image-to-4D (Animate124, Objaverse) | CLIP-I: 0.9372 -> 0.9505 | Improved |
| use DIMO for diverse motions instead of DG4D | DG4D · Image-to-4D (Animate124, Objaverse) | motion diversity: 10.2% -> 59.4% | Improved |
| use DIMO instead of 4D-fy for text-guided motion | 4D-fy · Text-to-4D (Animate124 examples) | motion amplitude: 0.254 -> 4.319 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- SV4D — SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistency (ICLR 2025) · better on 5 of 5 shared results
- STAG4D — STAG4D: Spatial-Temporal Anchored Generative 4D Gaussians (ECCV 2024) · better on 3 of 3 shared results
- Consistent4D — Consistent4D: Consistent 360° Dynamic Object Generation from Monocular Video (ICLR 2024) · 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
- Motion Diversification Networks (CVPR 2024 · similarity 0.88)
- Harmonizing Stochasticity and Determinism: Scene-responsive Diverse Human Motion Prediction (NeurIPS 2024 · similarity 0.87)
- World from Motion: Generative Dynamic Gaussian Reconstruction from Monocular Video (NeurIPS 2026 · similarity 0.87)
- FreeGave: 3D Physics Learning from Dynamic Videos by Gaussian Velocity (CVPR 2025 · similarity 0.86)
- DreamScene4D: Dynamic Multi-Object Scene Generation from Monocular Videos (NeurIPS 2024 · similarity 0.86)
- 3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation (ICLR 2025 · similarity 0.86)
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.
