United We Learn Better: Harvesting Learning Improvements From Class Hierarchies Across Tasks

TL;DR

This paper introduces a theoretical framework for leveraging class hierarchies in computer vision tasks like classification and detection, enabling hierarchical learning in sigmoid-based detection architectures.

Contribution

It develops a probabilistic and set-theoretic approach to extract parent predictions and design hierarchical loss functions applicable across different vision tasks.

Findings

Demonstrates improved hierarchical learning in classification benchmarks.

Shows applicability of hierarchical loss in object detection tasks.

Enables hierarchical learning in sigmoid-based detection architectures.

Abstract

Attempts of learning from hierarchical taxonomies in computer vision have been mostly focusing on image classification. Though ways of best harvesting learning improvements from hierarchies in classification are far from being solved, there is a need to target these problems in other vision tasks such as object detection. As progress on the classification side is often dependent on hierarchical cross-entropy losses, novel detection architectures using sigmoid as an output function instead of softmax cannot easily apply these advances, requiring novel methods in detection. In this work we establish a theoretical framework based on probability and set theory for extracting parent predictions and a hierarchical loss that can be used across tasks, showing results across classification and detection benchmarks and opening up the possibility of hierarchical learning for sigmoid-based detection architectures.