SA-FPN: An effective feature pyramid network for crowded human detection

• Published in: Applied intelligence (Dordrecht, Netherlands) Vol. 52; no. 11; pp. 12556 - 12568
• Main Authors: Zhou, Xinxin, Zhang, Long
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.09.2022; Springer Nature B.V
• Subjects: Artificial Intelligence; Computer Science; Detectors; Effectiveness; Feature maps; Human performance; Machines; Manufacturing; Mechanical Engineering; Modules; Object recognition; Occlusion; Performance enhancement; Processes; Sensors; Structural hierarchy; Crowd; Convolutional neural networks; Human detection; Feature pyramid networks; Object detection

The crowded scenario not only contains instances at various scales but also introduces a variety of occlusion patterns ranging from non-occluded situations to heavily occluded cases, making the shapes of the instances different. All of those can result in human detectors being hard to apply to them....