Horizontal Feature Pyramid Network for Object Detection in UAV Images

Abstract

Vision transformers have recently shown strong global context modeling capabilities in camouflaged object detection. However, they suffer from two major limitations: less effective locality modeling and insufficient feature aggregation in decoders, which are not conducive to camouflaged object detection that explores subtle cues from indistinguishable backgrounds. To address these issues, in this paper, we propose a novel transformer-based Feature Shrinkage Pyramid Network (FSPNet), which aims to hierarchically decode locality-enhanced neighboring transformer features through progressive shrinking for camouflaged object detection. Specifically, we propose a nonlocal token enhancement module (NL-TEM) that employs the non-local mechanism to interact neighboring tokens and explore graph-based high-order relations within tokens to enhance local representations of transformers. Moreover, we design a feature shrinkage decoder (FSD) with adjacent interaction modules (AIM), which progressively aggregates adjacent transformer features through a layer-bylayer shrinkage pyramid to accumulate imperceptible but effective cues as much as possible for object information decoding. Extensive quantitative and qualitative experiments demonstrate that the proposed model significantly outperforms the existing 24 competitors on three challenging COD benchmark datasets under six widely-used evaluation metrics. Our code is publicly available at https: //github.com/ZhouHuang23/FSPNet.

Remote Sensing Micro-Object Detection under Global and Local Attention Mechanism

“…In the Visdrone-2019 benchmark test, the method proposed in this study achieved 32.00% and 52.30% on the AP75 and AP50 metrics, respectively, which was significantly higher than other reference methods by at least 4.1% and 6.7%. Compared with Cas-cadeRCNN+ResNeXt [41] and TOOD+SF+SAHI+FI+PO [42] in remote sensing tiny object detection, our method showed superiority in all scenarios, especially in the AP50 metric representing overall performance. Additionally, compared with the latest EdgeYOLO [39], which incorporates enhanced data augmentation strategies, our method showed noticeable improvements in the APS and APM metrics, representing the effectiveness in tiny object detection.…”