LGADet: Light-weight Anchor-free Multispectral Pedestrian Detection with Mixed Local and Global Attention

Article Computer Science, Artificial Intelligence

FCOS: A Simple and Strong Anchor-Free Object Detector

Zhi Tian et al.

Summary: FCOS is a fully convolutional one-stage object detector that is anchor box free and achieves higher detection accuracy through post-processing non-maximum suppression.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2022)

Add to Collection

Article Computer Science, Information Systems

Bottom-up and Layerwise Domain Adaptation for Pedestrian Detection in Thermal Images

My Kieu et al.

Summary: This article explores domain adaptation approaches to adapt RGB-trained detectors to the thermal domain for pedestrian detection. Experimental evaluation shows that their bottom-up domain adaptation techniques outperform the best-performing single-modality pedestrian detection results on KAIST and outperform the state of the art on FLIR. The study proposes two new bottom-up domain adaptation strategies and highlights the importance of adapting detectors to variable lighting conditions for safety and security applications.

ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS (2021)

Add to Collection

Article Robotics

MLPD: Multi-Label Pedestrian Detector in Multispectral Domain

Jiwon Kim et al.

Summary: This study introduces a novel single-stage detection framework for multispectral pedestrian detection, which leverages multi-label learning to learn input state-aware features, and proposes a novel augmentation strategy to handle unpaired multispectral images.

IEEE ROBOTICS AND AUTOMATION LETTERS (2021)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Guided Attentive Feature Fusion for Multispectral Pedestrian Detection

Heng Zhang et al.

Summary: The proposed novel attentive multispectral feature fusion approach utilizes deep learning architecture to dynamically weigh and fuse multispectral features guided by inter- and intra-modality attention modules, significantly improving pedestrian detection accuracy at a low computation cost based on experiments on two public datasets.

2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2021) (2021)

Add to Collection

Proceedings Paper Computer Science, Software Engineering