BriefGPT.xyz
May, 2021
自监督多帧单眼场景流
Self-Supervised Multi-Frame Monocular Scene Flow
HTML
PDF
Junhwa Hur, Stefan Roth
TL;DR
本文通过自监督学习介绍了一种基于多帧单目场景流网络的方法，该方法在保持实时效率的同时显著提高了准确性，并通过在KITTI数据集上进行测试达到了自我监督学习中的最新水平。
Abstract
Estimating
3d scene flow
from a sequence of
monocular images
has been gaining increased attention due to the simple, economical capture setup. Owing to the severe ill-posedness of the problem, the accuracy of cur
→