On Hyperbolic Embeddings in 2D Object Detection

TL;DR

This paper investigates the use of hyperbolic geometry for object detection, demonstrating that it better captures class hierarchies and improves detection accuracy across various architectures and benchmarks.

Contribution

It introduces hyperbolic classifiers into object detection models and shows they enhance performance by aligning with the inherent class hierarchy structure.

Findings

Hyperbolic embeddings reveal categorical class hierarchies.

Hyperbolic classifiers reduce classification errors.

Overall detection performance is improved across architectures.

Abstract

Object detection, for the most part, has been formulated in the euclidean space, where euclidean or spherical geodesic distances measure the similarity of an image region to an object class prototype. In this work, we study whether a hyperbolic geometry better matches the underlying structure of the object classification space. We incorporate a hyperbolic classifier in two-stage, keypoint-based, and transformer-based object detection architectures and evaluate them on large-scale, long-tailed, and zero-shot object detection benchmarks. In our extensive experimental evaluations, we observe categorical class hierarchies emerging in the structure of the classification space, resulting in lower classification errors and boosting the overall object detection performance.