Saved in:
Bibliographic Details
Main Authors: Xu, Yingjie, Liu, Bangzhen, Tang, Hao, Deng, Bailin, He, Shengfeng
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.17638
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911817062678528
author Xu, Yingjie
Liu, Bangzhen
Tang, Hao
Deng, Bailin
He, Shengfeng
author_facet Xu, Yingjie
Liu, Bangzhen
Tang, Hao
Deng, Bailin
He, Shengfeng
contents We propose a voxel-based optimization framework, ReVoRF, for few-shot radiance fields that strategically address the unreliability in pseudo novel view synthesis. Our method pivots on the insight that relative depth relationships within neighboring regions are more reliable than the absolute color values in disoccluded areas. Consequently, we devise a bilateral geometric consistency loss that carefully navigates the trade-off between color fidelity and geometric accuracy in the context of depth consistency for uncertain regions. Moreover, we present a reliability-guided learning strategy to discern and utilize the variable quality across synthesized views, complemented by a reliability-aware voxel smoothing algorithm that smoothens the transition between reliable and unreliable data patches. Our approach allows for a more nuanced use of all available data, promoting enhanced learning from regions previously considered unsuitable for high-quality reconstruction. Extensive experiments across diverse datasets reveal that our approach attains significant gains in efficiency and accuracy, delivering rendering speeds of 3 FPS, 7 mins to train a $360^\circ$ scene, and a 5\% improvement in PSNR over existing few-shot methods. Code is available at https://github.com/HKCLynn/ReVoRF.
format Preprint
id arxiv_https___arxiv_org_abs_2403_17638
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Learning with Unreliability: Fast Few-shot Voxel Radiance Fields with Relative Geometric Consistency
Xu, Yingjie
Liu, Bangzhen
Tang, Hao
Deng, Bailin
He, Shengfeng
Computer Vision and Pattern Recognition
We propose a voxel-based optimization framework, ReVoRF, for few-shot radiance fields that strategically address the unreliability in pseudo novel view synthesis. Our method pivots on the insight that relative depth relationships within neighboring regions are more reliable than the absolute color values in disoccluded areas. Consequently, we devise a bilateral geometric consistency loss that carefully navigates the trade-off between color fidelity and geometric accuracy in the context of depth consistency for uncertain regions. Moreover, we present a reliability-guided learning strategy to discern and utilize the variable quality across synthesized views, complemented by a reliability-aware voxel smoothing algorithm that smoothens the transition between reliable and unreliable data patches. Our approach allows for a more nuanced use of all available data, promoting enhanced learning from regions previously considered unsuitable for high-quality reconstruction. Extensive experiments across diverse datasets reveal that our approach attains significant gains in efficiency and accuracy, delivering rendering speeds of 3 FPS, 7 mins to train a $360^\circ$ scene, and a 5\% improvement in PSNR over existing few-shot methods. Code is available at https://github.com/HKCLynn/ReVoRF.
title Learning with Unreliability: Fast Few-shot Voxel Radiance Fields with Relative Geometric Consistency
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.17638