Self-supervised Segmentation via Background Inpainting
This addresses the challenge of expensive data annotation for object segmentation in novel visual domains, though it is incremental as it builds on self-supervised and proposal-based methods.
The paper tackles the problem of poor generalization in supervised object detection and segmentation to visually different images by introducing a self-supervised approach that uses background inpainting and a Monte Carlo training strategy. It outperforms existing self-supervised methods in human detection and segmentation on non-standard images.
While supervised object detection and segmentation methods achieve impressive accuracy, they generalize poorly to images whose appearance significantly differs from the data they have been trained on. To address this when annotating data is prohibitively expensive, we introduce a self-supervised detection and segmentation approach that can work with single images captured by a potentially moving camera. At the heart of our approach lies the observation that object segmentation and background reconstruction are linked tasks, and that, for structured scenes, background regions can be re-synthesized from their surroundings, whereas regions depicting the moving object cannot. We encode this intuition into a self-supervised loss function that we exploit to train a proposal-based segmentation network. To account for the discrete nature of the proposals, we develop a Monte Carlo-based training strategy that allows the algorithm to explore the large space of object proposals. We apply our method to human detection and segmentation in images that visually depart from those of standard benchmarks and outperform existing self-supervised methods.