Self-Supervised Learning of Object Segmentation from Unlabeled RGB-D Videos
This addresses the problem of object segmentation for robotics and computer vision applications without requiring manual labeling, though it is incremental as it builds on existing self-supervised and graph-matching techniques.
This work tackles the problem of segmenting rigid objects in RGB images by developing a self-supervised learning system trained on unlabeled RGB-D videos, achieving superior performance over existing unsupervised methods on cluttered tabletop object datasets.
This work proposes a self-supervised learning system for segmenting rigid objects in RGB images. The proposed pipeline is trained on unlabeled RGB-D videos of static objects, which can be captured with a camera carried by a mobile robot. A key feature of the self-supervised training process is a graph-matching algorithm that operates on the over-segmentation output of the point cloud that is reconstructed from each video. The graph matching, along with point cloud registration, is able to find reoccurring object patterns across videos and combine them into 3D object pseudo labels, even under occlusions or different viewing angles. Projected 2D object masks from 3D pseudo labels are used to train a pixel-wise feature extractor through contrastive learning. During online inference, a clustering method uses the learned features to cluster foreground pixels into object segments. Experiments highlight the method's effectiveness on both real and synthetic video datasets, which include cluttered scenes of tabletop objects. The proposed method outperforms existing unsupervised methods for object segmentation by a large margin.