Improving Few-Shot Part Segmentation using Coarse Supervision
This addresses the annotation bottleneck for part segmentation in computer vision, offering a practical solution for domains with limited detailed labels, though it is incremental in leveraging existing coarse annotations.
The paper tackles the problem of high annotation cost for part segmentation by proposing a framework that leverages coarse labels like figure-ground masks and keypoints to improve models, achieving performance gains over baselines on benchmarks like Caltech-UCSD birds and OID Aircraft.
A significant bottleneck in training deep networks for part segmentation is the cost of obtaining detailed annotations. We propose a framework to exploit coarse labels such as figure-ground masks and keypoint locations that are readily available for some categories to improve part segmentation models. A key challenge is that these annotations were collected for different tasks and with different labeling styles and cannot be readily mapped to the part labels. To this end, we propose to jointly learn the dependencies between labeling styles and the part segmentation model, allowing us to utilize supervision from diverse labels. To evaluate our approach we develop a benchmark on the Caltech-UCSD birds and OID Aircraft dataset. Our approach outperforms baselines based on multi-task learning, semi-supervised learning, and competitive methods relying on loss functions manually designed to exploit sparse-supervision.