Digging Into Self-Supervised Monocular Depth Estimation
Per-pixel ground-truth depth data is challenging to acquire at scale. To overcome this limitation, self-supervised learning has emerged as a promising alternative for training models to perform monocular depth estimation. In this paper, we propose a set of improvements, which together result in both quantitatively and qualitatively improved depth maps compared to competing self-supervised methods. Research on self-supervised monocular training usually explores increasingly complex architectures, loss functions, and image formation models, all of which have recently helped to close the gap with fully-supervised methods. We show that a surprisingly simple model, and associated design choices, lead to superior predictions. In particular, we propose (i) a minimum reprojection loss, designed to robustly handle occlusions, (ii) a full-resolution multi-scale sampling method that reduces visual artifacts, and (iii) an auto-masking loss to ignore training pixels that violate camera motion assumptions. We demonstrate the effectiveness of each component in isolation, and show high quality, state-of-the-art results on the KITTI benchmark.
Code (15)
Tasks
Camera Pose EstimationDepth EstimationImage ReconstructionMonocular Depth EstimationMotion EstimationScene UnderstandingSelf-Supervised LearningUnsupervised Monocular Depth EstimationSimilar Papers 제목 키워드 기반
Digging Into Self-Supervised Monocular Depth Estimation
Per-pixel ground-truth depth data is challenging to acquire at scale. To overcome this limitation, self-supervised learning has emerged as a promising alternative for training models to perform monocular depth estimation…
Depth EstimationMonocular Depth EstimationSelf-Supervised LearningDeep Digging into the Generalization of Self-Supervised Monocular Depth Estimation
Self-supervised monocular depth estimation has been widely studied recently. Most of the work has focused on improving performance on benchmark datasets, such as KITTI, but has offered a few experiments on generalization…
Depth EstimationMonocular Depth EstimationSelfTune: Metrically Scaled Monocular Depth Estimation through Self-Supervised Learning
Monocular depth estimation in the wild inherently predicts depth up to an unknown scale. To resolve scale ambiguity issue, we present a learning algorithm that leverages monocular simultaneous localization and mapping (S…
Depth EstimationMonocular Depth EstimationRobot NavigationSelf-Supervised Learning+1Self-Supervised Monocular Scene Decomposition and Depth Estimation
Self-supervised monocular depth estimation approaches either ignore independently moving objects in the scene or need a separate segmentation step to identify them. We propose MonoDepthSeg to jointly estimate depth and s…
Depth EstimationMonocular Depth EstimationDigging Into Output Representation for Monocular 3D Object Detection
Monocular 3D object detection aims to recognize and localize objects in 3D space from a single image. Recent researches have conducted remarkable advancements, while all of them follow a typical output representation in…
3D Object DetectionMonocular 3D Object DetectionObjectobject-detection+1