Feature Weighting and Boosting for Few-Shot Segmentation
It addresses the problem of segmenting objects with limited training data for computer vision applications, but appears incremental as it builds on existing few-shot segmentation methods.
The paper tackles few-shot segmentation of foreground objects in images by improving feature discriminativeness and using an ensemble for inference, resulting in significant performance gains on PASCAL-$5^i$ and COCO-$20^i$ datasets.
This paper is about few-shot segmentation of foreground objects in images. We train a CNN on small subsets of training images, each mimicking the few-shot setting. In each subset, one image serves as the query and the other(s) as support image(s) with ground-truth segmentation. The CNN first extracts feature maps from the query and support images. Then, a class feature vector is computed as an average of the support's feature maps over the known foreground. Finally, the target object is segmented in the query image by using a cosine similarity between the class feature vector and the query's feature map. We make two contributions by: (1) Improving discriminativeness of features so their activations are high on the foreground and low elsewhere; and (2) Boosting inference with an ensemble of experts guided with the gradient of loss incurred when segmenting the support images in testing. Our evaluations on the PASCAL-$5^i$ and COCO-$20^i$ datasets demonstrate that we significantly outperform existing approaches.