Group cost-sensitive boosting with multi-scale decorrelated filters for pedestrian detection

We propose a novel two-stage pedestrian detection framework that combines multiscale decorrelated filters to extract more discriminative features and a novel group costsensitive boosting algorithm. The proposed boosting algorithm is based on mixture loss to alleviate the influence of annotation erro...

Full description

Bibliographic Details
Main Authors: Zhou, Chengju, Wu, Meiqing, Lam, Siew-Kei
Other Authors: School of Computer Science and Engineering
Format: Conference Paper
Language:English
Published: 2021
Subjects:
Online Access:https://hdl.handle.net/10356/147489