Neural Field Classifiers via Target Encoding and Classification Loss

arXiv:2403.01058v1 Announce Type: cross
Abstract: Neural field methods have seen great progress in various long-standing tasks in computer vision and computer graphics, including novel view synthesis and geometry reconstruction. As existing neural field methods try to predict some coordinate-based continuous target values, such as RGB for Neural Radiance Field (NeRF), all of these methods are regression models and are optimized by some regression loss. However, are regression models really better than classification models for neural field methods? In this work, …

abstract arxiv classification classifiers computer computer graphics computer vision continuous cs.cv cs.lg encoding geometry graphics loss nerf neural radiance field novel progress synthesis tasks type values via view vision