AugFPN: Improving Multi-scale Feature Learning for Object Detection

TL;DR

AugFPN introduces a novel feature pyramid architecture that enhances multi-scale feature learning for object detection, addressing FPN's limitations and improving detection accuracy across various models.

Contribution

The paper proposes AugFPN, a new feature pyramid architecture with three components, to better exploit multi-scale features and improve object detection performance.

Findings

Achieves 2.3 and 1.6 AP improvements in Faster R-CNN with ResNet50 and MobileNet-v2.

Enhances RetinaNet by 1.6 AP and FCOS by 0.9 AP with ResNet50.

Addresses FPN's design defects to fully exploit multi-scale features.

Abstract

Current state-of-the-art detectors typically exploit feature pyramid to detect objects at different scales. Among them, FPN is one of the representative works that build a feature pyramid by multi-scale features summation. However, the design defects behind prevent the multi-scale features from being fully exploited. In this paper, we begin by first analyzing the design defects of feature pyramid in FPN, and then introduce a new feature pyramid architecture named AugFPN to address these problems. Specifically, AugFPN consists of three components: Consistent Supervision, Residual Feature Augmentation, and Soft RoI Selection. AugFPN narrows the semantic gaps between features of different scales before feature fusion through Consistent Supervision. In feature fusion, ratio-invariant context information is extracted by Residual Feature Augmentation to reduce the information loss of feature map at the highest pyramid level. Finally, Soft RoI Selection is employed to learn a better RoI feature adaptively after feature fusion. By replacing FPN with AugFPN in Faster R-CNN, our models achieve 2.3 and 1.6 points higher Average Precision (AP) when using ResNet50 and MobileNet-v2 as backbone respectively. Furthermore, AugFPN improves RetinaNet by 1.6 points AP and FCOS by 0.9 points AP when using ResNet50 as backbone. Codes will be made available.