Papers › Attention-guided Context Feature Pyramid Network for Object Detection

Attention-guided Context Feature Pyramid Network for Object Detection

23 May 2020arXiv:2005.11475archive 2025-07-28

Junxu Cao, Qi Chen, Jun Guo, Ruichao Shi

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.

PaperPDFCode

Code

Caojunxu/AC-FPN officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Instance SegmentationObjectObject DetectionOpen-Ended Question AnsweringSemantic Segmentationobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Object Detection COCO test-dev AC-FPN Cascade R-CNN (X-152-32x8d-FPN-IN5k, multi scale, only CEM) AP50 70.4 #76 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN (X-152-32x8d-FPN-IN5k, multi scale, only CEM) AP75 57 #76 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN (X-152-32x8d-FPN-IN5k, multi scale, only CEM) APL 64.7 #76 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN (X-152-32x8d-FPN-IN5k, multi scale, only CEM) APM 54.8 #76 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN (X-152-32x8d-FPN-IN5k, multi scale, only CEM) APS 34.2 #76 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN (X-152-32x8d-FPN-IN5k, multi scale, only CEM) box mAP 51.9 #76 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN(ResNet-101, single scale) AP50 64.4 #139 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN(ResNet-101, single scale) AP75 49 #139 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN(ResNet-101, single scale) APL 56.6 #139 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN(ResNet-101, single scale) APM 47.7 #139 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN(ResNet-101, single scale) APS 26.9 #139 of 225 Archive leaderboard report
Object Detection COCO test-dev AC-FPN Cascade R-CNN(ResNet-101, single scale) box mAP 45 #139 of 225 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AM

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections