Fine-grained domain adaptive crowd counting via point-derived segmentation

• Main Authors: LIU, Yongtuo, XU, Dan, REN, Sucheng, WU, Hanjie, CAI, Hongmin, HE, Shengfeng
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2023
• Subjects: Crowd counting; Domain adaptation; Pointderived segmentation; Computer vision; Artificial Intelligence and Robotics; Graphics and Human Computer Interfaces
• Online Access: https://ink.library.smu.edu.sg/sis_research/8443; https://ink.library.smu.edu.sg/context/sis_research/article/9446/viewcontent/2108.02980.pdf
• Institution: Singapore Management University

Due to domain shift, a large performance drop is usually observed when a trained crowd counting model is deployed in the wild. While existing domain-adaptive crowd counting methods achieve promising results, they typically regard each crowd image as a whole and reduce domain discrepancies in a holis...