Relation Networks for Object Detection

TL;DR

This paper introduces a lightweight object relation module that models interactions between objects in deep learning-based detection, improving recognition and enabling the first fully end-to-end object detector.

Contribution

It proposes a novel, lightweight, and easy-to-embed relation module that models object interactions, enhancing CNN-based detection without extra supervision.

Findings

Improves object recognition accuracy

Enhances duplicate removal in detection pipelines

Enables fully end-to-end object detection

Abstract

Although it is well believed for years that modeling relations between objects would help object recognition, there has not been evidence that the idea is working in the deep learning era. All state-of-the-art object detection systems still rely on recognizing object instances individually, without exploiting their relations during learning. This work proposes an object relation module. It processes a set of objects simultaneously through interaction between their appearance feature and geometry, thus allowing modeling of their relations. It is lightweight and in-place. It does not require additional supervision and is easy to embed in existing networks. It is shown effective on improving object recognition and duplicate removal steps in the modern object detection pipeline. It verifies the efficacy of modeling object relations in CNN based detection. It gives rise to the first fully end-to-end object detector.