AFPN: Asymptotic Feature Pyramid Network for Object Detection

TL;DR

AFPN introduces an asymptotic feature fusion method that enhances multi-scale object detection by directly connecting non-adjacent feature levels, improving semantic integration and detection accuracy.

Contribution

The paper proposes AFPN, a novel feature pyramid network that supports direct non-adjacent level interaction and adaptive spatial fusion, addressing information loss in traditional methods.

Findings

Achieves superior detection performance on MS-COCO dataset.

Effectively mitigates feature information loss and conflicts during fusion.

Outperforms existing state-of-the-art feature pyramid networks.

Abstract

Multi-scale features are of great importance in encoding objects with scale variance in object detection tasks. A common strategy for multi-scale feature extraction is adopting the classic top-down and bottom-up feature pyramid networks. However, these approaches suffer from the loss or degradation of feature information, impairing the fusion effect of non-adjacent levels. This paper proposes an asymptotic feature pyramid network (AFPN) to support direct interaction at non-adjacent levels. AFPN is initiated by fusing two adjacent low-level features and asymptotically incorporates higher-level features into the fusion process. In this way, the larger semantic gap between non-adjacent levels can be avoided. Given the potential for multi-object information conflicts to arise during feature fusion at each spatial location, adaptive spatial fusion operation is further utilized to mitigate these inconsistencies. We incorporate the proposed AFPN into both two-stage and one-stage object detection frameworks and evaluate with the MS-COCO 2017 validation and test datasets. Experimental evaluation shows that our method achieves more competitive results than other state-of-the-art feature pyramid networks. The code is available at \href{https://github.com/gyyang23/AFPN}{https://github.com/gyyang23/AFPN}.