Research map / 3D vision / 3D generation
OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use CurveNet rather than the other tested point-cloud architectures | Point-cloud classification trained on ModelNet-40 · OmniObject3D OOD-style robustness | OA_Style: 0.500 | Helped |
| Use GDANet rather than the other tested point-cloud architectures | Point-cloud classification trained on ModelNet-40 · OmniObject3D-C | mCE: 0.920 | Helped |
| Use CurveNet rather than SimpleView | Point-cloud classification trained on ModelNet-40 · OmniObject3D OOD-style robustness | OA_Style: 0.476 -> 0.500 | Helped |
| Use Plenoxels rather than NeRF or mip-NeRF | Single-scene novel view synthesis · OmniObject3D | PSNR: 34.01 -> 41.04 | Helped |
| Use Plenoxels rather than NeRF or mip-NeRF | Single-scene novel view synthesis · OmniObject3D | LPIPS: 0.068 -> 0.030 | Helped |
| Train across categories excluding test categories instead of only on the test category | MVSNeRF · Cross-scene NVS, 10 categories | PSNR: 17.54 -> 17.49 | Mixed |
| Train across categories excluding test categories instead of only on the test category | IBRNet · Cross-scene NVS, 10 categories | PSNR: 19.03 -> 19.39 | Helped |
| Fine-tune on each test scene after cross-scene training | IBRNet · Cross-scene NVS, 10 categories | PSNR: 19.39 -> 26.89 | Helped |
| Use Voxurf rather than NeuS or VolSDF | Dense-view surface reconstruction · OmniObject3D | Chamfer Distance × 10^3: 6.09 -> 5.49 | Helped |
| Train SparseNeuS on all categories rather than one category | SparseNeuS · Sparse-view surface reconstruction, 3 views | Chamfer Distance × 10^3: 32.36 -> 25.00 | Helped |
| Train SparseNeuS on all categories rather than 10 categories | SparseNeuS · Sparse-view surface reconstruction, 3 views | Chamfer Distance × 10^3: 29.87 -> 25.00 | Helped |
| Train GET3D on the Fruits subset rather than Furniture | GET3D · 3D generation | MMD: 4.27 -> 3.32 | Helped |
Closest papers by meaning
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.90)
- EA3D: Online Open-World 3D Object Extraction from Streaming Videos (NeurIPS 2025 · similarity 0.89)
- ImOV3D: Learning Open Vocabulary Point Clouds 3D Object Detection from Only 2D Images (NeurIPS 2024 · similarity 0.88)
- UniK3D: Universal Camera Monocular 3D Estimation (CVPR 2025 · similarity 0.88)
- UnCommon Objects in 3D (CVPR 2025 · similarity 0.88)
- Large-Vocabulary 3D Diffusion Model with Transformer (ICLR 2024 · 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.