動的シーンにおける教師なし単眼深度学習
Unsupervised Monocular Depth Learning in Dynamic Scenes
単眼の光度整合性のみを教師信号として、深度・自己運動・物体の3D並進場を同時に学習する手法を提案。3D並進場の疎性と剛体物体での一定性という事前知識で正則化することで、動的シーンでも高精度な深度推定を実現した。
著者: Hanhan Li, Ariel Gordon, Hang Zhao, Vincent Casser, Anelia Angelova
分類: cs.CV, cs.GR, cs.LG, cs.RO
原文アブストラクト
We present a method for jointly training the estimation of depth, ego-motion, and a dense 3D translation field of objects relative to the scene, with monocular photometric consistency being the sole source of supervision. We show that this apparently heavily underdetermined problem can be regularized by imposing the following prior knowledge about 3D translation fields: they are sparse, since most of the scene is static, and they tend to be constant for rigid moving objects. We show that this regularization alone is sufficient to train monocular depth prediction models that exceed the accuracy achieved in prior work for dynamic scenes, including methods that require semantic input. Code is at https://github.com/google-research/google-research/tree/master/depth_and_motion_learning .
