The Pix3D dataset is a large-scale benchmark of diverse image-shape pairs with pixel-level 2D-3D alignment. Pix3D has wide applications in shape-related tasks including reconstruction, retrieval, viewpoint estimation, etc.
Source: http://pix3d.csail.mit.edu/
Image Source: http://pix3d.csail.mit.edu/
Variants: Pix3D S2, Pix3D S1, Pix3D
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Pose Estimation | Mid-Level based | Object Pose Estimation using Mid-level … | 2022-03-02 |
| 3D Shape Reconstruction | IM3D | Holistic 3D Scene Understanding from … | 2021-03-11 |
| 3D Shape Reconstruction | MGN | Total3DUnderstanding: Joint Layout, Object Pose … | 2020-02-27 |
| 3D Shape Reconstruction | TMN | Deep Mesh Reconstruction from Single … | 2019-09-01 |
| 3D Shape Reconstruction | MarrNet extension (w/ Pose) | Pix3D: Dataset and Methods for … | 2018-04-12 |
| 3D Shape Reconstruction | AtlasNet | AtlasNet: A Papier-Mâché Approach to … | 2018-02-15 |
Recent papers with results on this dataset: