2 Hits in 3.3 sec

CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis [article]

Peng Zhou, Lingxi Xie, Bingbing Ni, Qi Tian
2021 arXiv   pre-print
This paper presents CIPS-3D, a style-based, 3D-aware generator that is composed of a shallow NeRF network and a deep implicit neural representation (INR) network.  ...  Trained on raw, single-view images, CIPS-3D sets new records for 3D-aware image synthesis with an impressive FID of 6.97 for images at the 256×256 resolution on FFHQ.  ...  We hope that CIPS-3D will serve as a good base model for downstream tasks such as 3D-aware GAN inversion and 3D-aware image editing.  ... 
arXiv:2110.09788v1 fatcat:3kv4ftilsfbdnlxyl6kvkwznmi

StyleSDF: High-Resolution 3D-Consistent Image and Geometry Generation [article]

Roy Or-El and Xuan Luo and Mengyi Shan and Eli Shechtman and Jeong Joon Park and Ira Kemelmacher-Shlizerman
2021 arXiv   pre-print
Our method is trained on single-view RGB data only, and stands on the shoulders of StyleGAN2 for image generation, while solving two main challenges in 3D-aware GANs: 1) high-resolution, view-consistent  ...  We achieve this by merging a SDF-based 3D representation with a style-based 2D generator.  ...  NeurIPS, [76] Peng Zhou, Lingxi Xie, Bingbing Ni, and Qi Tian. 2019. 1, 2 Cips-3d: A 3d-aware generator of gans based on [64] Ayush  ... 
arXiv:2112.11427v1 fatcat:evrfhytuzbgifgksf62zjgzjmq