Multi-utility Learning: Structured-output Learning with Multiple Annotation-specific Loss Functions

TL;DR

This paper introduces a multi-utility learning framework for structured prediction that effectively leverages various forms of weak annotations, significantly improving semantic image segmentation accuracy.

Contribution

It proposes a unified method to infer annotation-specific loss functions, enabling learning from diverse supervision types in structured-output tasks.

Findings

Annotation-specific loss functions improve segmentation accuracy.

Framework effectively utilizes weak annotations like bounding boxes and image labels.

Significant performance gains over single-annotation baseline systems.

Abstract

Structured-output learning is a challenging problem; particularly so because of the difficulty in obtaining large datasets of fully labelled instances for training. In this paper we try to overcome this difficulty by presenting a multi-utility learning framework for structured prediction that can learn from training instances with different forms of supervision. We propose a unified technique for inferring the loss functions most suitable for quantifying the consistency of solutions with the given weak annotation. We demonstrate the effectiveness of our framework on the challenging semantic image segmentation problem for which a wide variety of annotations can be used. For instance, the popular training datasets for semantic segmentation are composed of images with hard-to-generate full pixel labellings, as well as images with easy-to-obtain weak annotations, such as bounding boxes around objects, or image-level labels that specify which object categories are present in an image. Experimental evaluation shows that the use of annotation-specific loss functions dramatically improves segmentation accuracy compared to the baseline system where only one type of weak annotation is used.