Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling. Wu, J., Zhang, C., Xue, T., Freeman, W., T., & Tenenbaum, J., B. Advances in Neural Information Processing Systems, 2016.
abstract   bibtex   
We study the problem of 3D object generation. We propose a novel framework, namely 3D Generative Adversarial Network (3D-GAN), which generates 3D objects from a probabilistic space by leveraging recent advances in volumetric convo-lutional networks and generative adversarial nets. The benefits of our model are three-fold: first, the use of an adversarial criterion, instead of traditional heuristic criteria, enables the generator to capture object structure implicitly and to synthesize high-quality 3D objects; second, the generator establishes a mapping from a low-dimensional probabilistic space to the space of 3D objects, so that we can sample objects without a reference image or CAD models, and explore the 3D object manifold; third, the adversarial discriminator provides a powerful 3D shape descriptor which, learned without supervision, has wide applications in 3D object recognition. Experiments demonstrate that our method generates high-quality 3D objects, and our unsupervisedly learned features achieve impressive performance on 3D object recognition, comparable with those of supervised learning methods.
@article{
 title = {Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling},
 type = {article},
 year = {2016},
 pages = {82-90},
 id = {88705a5a-78a1-37e1-a854-8d914409efe9},
 created = {2022-02-08T07:56:57.258Z},
 file_attached = {true},
 profile_id = {235249c2-3ed4-314a-b309-b1ea0330f5d9},
 group_id = {1ff583c0-be37-34fa-9c04-73c69437d354},
 last_modified = {2022-09-08T17:25:38.256Z},
 read = {true},
 starred = {false},
 authored = {false},
 confirmed = {true},
 hidden = {false},
 citation_key = {Wu2016},
 folder_uuids = {b6d75013-efe2-4ddc-b3db-65496bd4db9f,10e04504-7e21-4b84-9037-5a4431df1a8a,1853f94b-7af1-40fa-b068-4758e9a02bc4,a3bb779f-cc19-49e7-8c16-ed7f763ec870,465973dc-f47a-4093-b987-b885254a351b},
 private_publication = {false},
 abstract = {We study the problem of 3D object generation. We propose a novel framework, namely 3D Generative Adversarial Network (3D-GAN), which generates 3D objects from a probabilistic space by leveraging recent advances in volumetric convo-lutional networks and generative adversarial nets. The benefits of our model are three-fold: first, the use of an adversarial criterion, instead of traditional heuristic criteria, enables the generator to capture object structure implicitly and to synthesize high-quality 3D objects; second, the generator establishes a mapping from a low-dimensional probabilistic space to the space of 3D objects, so that we can sample objects without a reference image or CAD models, and explore the 3D object manifold; third, the adversarial discriminator provides a powerful 3D shape descriptor which, learned without supervision, has wide applications in 3D object recognition. Experiments demonstrate that our method generates high-quality 3D objects, and our unsupervisedly learned features achieve impressive performance on 3D object recognition, comparable with those of supervised learning methods.},
 bibtype = {article},
 author = {Wu, Jiajun and Zhang, Chengkai and Xue, Tianfan and Freeman, William T. and Tenenbaum, Joshua B.},
 journal = {Advances in Neural Information Processing Systems},
 number = {Nips}
}

Downloads: 0