Research map / Point clouds / Point cloud classification
Collaborative Multi-Modal Coding for High-Quality 3D Generation
Ziang Cao, Zhaoxi Chen, Liang Pan, 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 3D SDF supervision to 2D supervision (ablation) | multi-modal coding · Objaverse subset reconstruction | PSNR: 24.26 -> 28.12 | Improved |
| add 3D SDF supervision (ablation) | multi-modal coding · Objaverse subset reconstruction | Chamfer Distance: 0.015 -> 0.0048 | Improved |
| add VAE compression of triplanes (ablation) | TriMM diffusion · generation | CLIP score: 43.2 -> 52.1 | Improved |
| add VAE compression of triplanes (ablation) | TriMM diffusion · generation | Chamfer Distance: 0.108 -> 0.034 | Improved |
| add reconstruction loss in diffusion training (ablation) | TriMM · multi-modal ablation | CLIP score: 57.5 -> 64.8 | Improved |
| add reconstruction loss in diffusion training (ablation) | TriMM · multi-modal ablation | Chamfer Distance: 0.048 -> 0.015 | Improved |
| add point cloud modality to RGB (ablation) | RGB-only generator · multi-modal ablation | Chamfer Distance: 0.084 -> 0.021 | Improved |
| add RGBD and point cloud to RGB (ablation) | TriMM · multi-modal ablation | CLIP score: 58.3 -> 64.8 | Improved |
| combine RGBD and point cloud without RGB (ablation) | TriMM · multi-modal ablation | PSNR: 16.6 -> 14.2 | Hurt |
| add WildRGB-D additional data (ablation) | TriMM · additional data | CLIP score: 62.4 -> 64.9 | Improved |
| add WildRGB-D additional data (ablation) | TriMM · additional data | Chamfer Distance: 0.022 -> 0.0018 | Improved |
| increase triplane resolution (ablation) | TriMM · reconstruction | reconstruction quality | Mixed |
Closest papers by meaning
- TAMM: TriAdapter Multi-Modal Learning for 3D Shape Understanding (CVPR 2024 · similarity 0.88)
- Structured 3D Latents for Scalable and Versatile 3D Generation (CVPR 2025 · similarity 0.88)
- Lafite: A Generative Latent Field for 3D Native Texturing (CVPR 2026 · similarity 0.88)
- PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers (NeurIPS 2025 · similarity 0.87)
- MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation (CVPR 2024 · similarity 0.87)
- Sharp-It: A Multi-view to Multi-view Diffusion Model for 3D Synthesis and Manipulation (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.
