Research map / 3D vision / 3D generation
MGD:Mesh-guided Gaussians with Diffusion Priors for Dynamic Objects Reconstruction from Monocular RGB-D Video
Weixing Xie, Ying Ye, Xian Wu, Jintian Li, Bingchuan Li, Yanchen Lin, Junfeng Yao
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 the diffusion prior (ablation) | MGD · Ablation (three datasets, avg) | CLIP sim.: 83.07 -> 87.96 | Improved |
| add the diffusion prior (ablation) | MGD · Ablation (three datasets, avg) | Acc. [cm]: 1.03 -> 0.86 | Improved |
| add depth ControlNet to the diffusion prior (ablation) | MGD · Ablation (three datasets, avg) | Acc. [cm]: 0.95 -> 0.86 | Improved |
| add scene-adaptive LoRA fine-tuning of the diffusion prior (ablation) | MGD · Ablation (three datasets, avg) | CLIP sim.: 87.29 -> 87.96 | Improved |
| add Global Gaussian Deformation Module (GGDM) (ablation) | MGD · Ablation (three datasets, avg) | Acc. [cm]: 1.10 -> 0.86 | Improved |
| add Local Gaussian Adjustment Module (LGAM) (ablation) | MGD · Ablation (three datasets, avg) | PSNR [dB]: 28.01 -> 29.52 | Improved |
| use MGD instead of MorpheuS (NeRF-based) | dynamic object reconstruction · Real-world datasets avg | PSNR [dB]: 26.44 -> 29.52 | Improved |
| use MGD instead of MorpheuS for training and rendering speed | dynamic object reconstruction · Real-world datasets | Train time / FPS: ~12h -> ~1h; 1.01 -> 35.26 FPS | Improved |
| use MGD instead of 4DTAM (Gaussians with TSDF mesh) | observed-view reconstruction · DeepDeform | PSNR [dB]: 24.15 -> 27.58 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MorpheuS — MorpheuS: Neural Dynamic 360deg Surface Reconstruction from Monocular RGB-D Video (CVPR 2024) · better on 6 of 6 shared results
- 4DTAM — 4DTAM: Non-Rigid Tracking and Mapping via Dynamic Surface Gaussians (CVPR 2025) · better on 4 of 4 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
- Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction (CVPR 2024 · similarity 0.89)
- MorpheuS: Neural Dynamic 360deg Surface Reconstruction from Monocular RGB-D Video (CVPR 2024 · similarity 0.88)
- Gaussian-Flow: 4D Reconstruction with Dynamic 3D Gaussian Particle (CVPR 2024 · similarity 0.88)
- Coherent 3D Scene Diffusion From a Single RGB Image (NeurIPS 2024 · similarity 0.88)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.88)
- DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videos (NeurIPS 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.
