Neural Field Classifiers via Target Encoding and Classification Loss

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Xindi, Xie, Zeke, Zhou, Xiong, Liu, Boyu, Liu, Buhua, Liu, Yi, Wang, Haoran, Cai, Yunfeng, Sun, Mingming
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929262479540224
author Yang, Xindi
Xie, Zeke
Zhou, Xiong
Liu, Boyu
Liu, Buhua
Liu, Yi
Wang, Haoran
Cai, Yunfeng
Sun, Mingming
author_facet Yang, Xindi
Xie, Zeke
Zhou, Xiong
Liu, Boyu
Liu, Buhua
Liu, Yi
Wang, Haoran
Cai, Yunfeng
Sun, Mingming
contents 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, we try to visit this very fundamental but overlooked question for neural fields from a machine learning perspective. We successfully propose a novel Neural Field Classifier (NFC) framework which formulates existing neural field methods as classification tasks rather than regression tasks. The proposed NFC can easily transform arbitrary Neural Field Regressor (NFR) into its classification variant via employing a novel Target Encoding module and optimizing a classification loss. By encoding a continuous regression target into a high-dimensional discrete encoding, we naturally formulate a multi-label classification task. Extensive experiments demonstrate the impressive effectiveness of NFC at the nearly free extra computational costs. Moreover, NFC also shows robustness to sparse inputs, corrupted images, and dynamic scenes.
format Preprint
id arxiv_https___arxiv_org_abs_2403_01058
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Neural Field Classifiers via Target Encoding and Classification Loss
Yang, Xindi
Xie, Zeke
Zhou, Xiong
Liu, Boyu
Liu, Buhua
Liu, Yi
Wang, Haoran
Cai, Yunfeng
Sun, Mingming
Computer Vision and Pattern Recognition
Machine Learning
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, we try to visit this very fundamental but overlooked question for neural fields from a machine learning perspective. We successfully propose a novel Neural Field Classifier (NFC) framework which formulates existing neural field methods as classification tasks rather than regression tasks. The proposed NFC can easily transform arbitrary Neural Field Regressor (NFR) into its classification variant via employing a novel Target Encoding module and optimizing a classification loss. By encoding a continuous regression target into a high-dimensional discrete encoding, we naturally formulate a multi-label classification task. Extensive experiments demonstrate the impressive effectiveness of NFC at the nearly free extra computational costs. Moreover, NFC also shows robustness to sparse inputs, corrupted images, and dynamic scenes.
title Neural Field Classifiers via Target Encoding and Classification Loss
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2403.01058