We lifted 11 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| elliottwu/unsup3d | canonical | 2 of 11 |
| Function | Status | Where it lives |
|---|---|---|
| make_dataset | Ran | elliottwu/unsup3d/unsup3d/dataloaders.py code served (permissive licence) · get_code("641352b2fce0b79c") |
| setup_runtime | Ran | elliottwu/unsup3d/unsup3d/utils.py code served (permissive licence) · get_code("cb8259004b33f483") |
| dump_yaml | Not yet run | elliottwu/unsup3d/unsup3d/utils.py code served (permissive licence) · get_code("fa8f0dbdb9969655") |
| export_to_obj_string | Not yet run | elliottwu/unsup3d/demo/utils.py code served (permissive licence) · get_code("7d8a940bdc2312d1") |
| get_data_loaders | Not yet run | elliottwu/unsup3d/unsup3d/dataloaders.py code served (permissive licence) · get_code("e49ea94e0987a6e8") |
| get_grid | Not yet run | elliottwu/unsup3d/demo/utils.py code served (permissive licence) · get_code("9a90a24d56749586") |
| is_image_file | Not yet run | elliottwu/unsup3d/unsup3d/dataloaders.py code served (permissive licence) · get_code("521cedf170be4833") |
| load_yaml | Not yet run | elliottwu/unsup3d/unsup3d/utils.py code served (permissive licence) · get_code("c9d96366dfea8bc9") |
| mm_normalize | Not yet run | elliottwu/unsup3d/unsup3d/renderer/utils.py code served (permissive licence) · get_code("0c5ab402075cd7d5") |
| rand_posneg_range | Not yet run | elliottwu/unsup3d/unsup3d/renderer/utils.py code served (permissive licence) · get_code("93ba7714c2ce3acc") |
| rand_range | Not yet run | elliottwu/unsup3d/unsup3d/renderer/utils.py code served (permissive licence) · get_code("17ad8150640684d6") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
We propose a method to learn 3D deformable object categories from raw single-view images, without external supervision. The method is based on an autoencoder that factors each input image into depth, albedo, viewpoint and illumination. In order to disentangle these components without supervision, we use the fact that many object categories have, at least in principle, a symmetric structure. We show that reasoning about illumination allows us to exploit the underlying object symmetry even if the appearance is not symmetric due to shading. Furthermore, we model objects that are probably, but not certainly, symmetric by predicting a symmetry probability map, learned end-to-end with the other components of the model. Our experiments show that this method can recover very accurately the 3D shape of human faces, cat faces and cars from single-view images, without any supervision or a prior shape model. On benchmarks, we demonstrate superior accuracy compared to another method that uses supervision at the level of 2D image correspondences.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("1911.11130")
get_code_for_paper("1911.11130")
have("1911.11130")
Connect an agent — have() is free.