自己教師あり単眼深度推定のための適応的離散視差ボリューム
Adaptive Discrete Disparity Volume for Self-supervised Monocular Depth Estimation
RGB画像ごとに深度分布を動的に感知して適応的なビンを生成する学習可能なモジュールADDVを提案し、自己教師あり単眼深度推定の精度を向上させた。
著者: Jianwei Ren
分類: cs.CV, cs.AI, cs.RO
原文アブストラクト
In self-supervised monocular depth estimation tasks, discrete disparity prediction has been proven to attain higher quality depth maps than common continuous methods. However, current discretization strategies often divide depth ranges of scenes into bins in a handcrafted and rigid manner, limiting model performance. In this paper, we propose a learnable module, Adaptive Discrete Disparity Volume (ADDV), which is capable of dynamically sensing depth distributions in different RGB images and generating adaptive bins for them. Without any extra supervision, this module can be integrated into existing CNN architectures, allowing networks to produce representative values for bins and a probability volume over them. Furthermore, we introduce novel training strategies - uniformizing and sharpening - through a loss term and temperature parameter, respectively, to provide regularizations under self-supervised conditions, preventing model degradation or collapse. Empirical results demonstrate that ADDV effectively processes global information, generating appropriate bins for various scenes and producing higher quality depth maps compared to handcrafted methods.
