Relation Distillation Networks for Video Object Detection

TL;DR

This paper introduces Relation Distillation Networks (RDN), a novel architecture that models and distills object relations over time to improve video object detection accuracy.

Contribution

The paper proposes RDN, a new method that aggregates and refines object relations across frames, enhancing detection and linking in videos.

Findings

Achieves state-of-the-art mAP on ImageNet VID dataset.

Improves object detection accuracy with relation modeling.

Outperforms existing methods with 83.8% and 84.7% mAP.

Abstract

It has been well recognized that modeling object-to-object relations would be helpful for object detection. Nevertheless, the problem is not trivial especially when exploring the interactions between objects to boost video object detectors. The difficulty originates from the aspect that reliable object relations in a video should depend on not only the objects in the present frame but also all the supportive objects extracted over a long range span of the video. In this paper, we introduce a new design to capture the interactions across the objects in spatio-temporal context. Specifically, we present Relation Distillation Networks (RDN) --- a new architecture that novelly aggregates and propagates object relation to augment object features for detection. Technically, object proposals are first generated via Region Proposal Networks (RPN). RDN then, on one hand, models object relation via multi-stage reasoning, and on the other, progressively distills relation through refining supportive object proposals with high objectness scores in a cascaded manner. The learnt relation verifies the efficacy on both improving object detection in each frame and box linking across frames. Extensive experiments are conducted on ImageNet VID dataset, and superior results are reported when comparing to state-of-the-art methods. More remarkably, our RDN achieves 81.8% and 83.2% mAP with ResNet-101 and ResNeXt-101, respectively. When further equipped with linking and rescoring, we obtain to-date the best reported mAP of 83.8% and 84.7%.