April 22, 2024 (v1)
Conference paper
Recent work has demonstrated that Generative Adversarial Networks (GANs) can be trained to generate 3D content from2D image collections, by synthesizing features for neural radiance field rendering. However, most such solutions generateradiance, with lighting entangled with materials. This results in unrealistic appearance, since lighting...
Uploaded on: March 9, 2024