Attention-guided Context Feature Pyramid Network for Object Detection

TL;DR

This paper introduces AC-FPN, a novel architecture that enhances object detection by integrating attention-guided multi-path features to effectively utilize large receptive fields and improve localization and recognition accuracy.

Contribution

The paper proposes a new Attention-guided Context Feature Pyramid Network (AC-FPN) that incorporates two modules, CEM and AM, to better exploit contextual information and salient dependencies in object detection.

Findings

Achieves state-of-the-art results on object detection and instance segmentation tasks.

Significantly improves localization and recognition accuracy over baseline models.

Easily integrates into existing FPN-based models.

Abstract

For object detection, how to address the contradictory requirement between feature map resolution and receptive field on high-resolution inputs still remains an open question. In this paper, to tackle this issue, we build a novel architecture, called Attention-guided Context Feature Pyramid Network (AC-FPN), that exploits discriminative information from various large receptive fields via integrating attention-guided multi-path features. The model contains two modules. The first one is Context Extraction Module (CEM) that explores large contextual information from multiple receptive fields. As redundant contextual relations may mislead localization and recognition, we also design the second module named Attention-guided Module (AM), which can adaptively capture the salient dependencies over objects by using the attention mechanism. AM consists of two sub-modules, i.e., Context Attention Module (CxAM) and Content Attention Module (CnAM), which focus on capturing discriminative semantics and locating precise positions, respectively. Most importantly, our AC-FPN can be readily plugged into existing FPN-based models. Extensive experiments on object detection and instance segmentation show that existing models with our proposed CEM and AM significantly surpass their counterparts without them, and our model successfully obtains state-of-the-art results. We have released the source code at https://github.com/Caojunxu/AC-FPN.