活动地点： 延长校区电机楼125
活动时间： 2017年9月1日 上午10:30
报告题目：教会计算机理解图像的关键技术
报告人：Associate Prof. Jien Kato,Nagoya University,Japan
内容提要：
This talk focuses on image/movie recognition in an approach of machine learning. The speaker will review main tasks and significant progress in this field. She will also introduce her own work in fine-grained classification of pedestrian. The purpose of this research is to classify sex (2 classes), age (5 classes), weight (3 classes), clothing (4 classes) subcategories of pedestrian. The approach called four add-on strategies is proposed. They are (1) super-resolution based image preprocessing, which helps to recover the image details; (2) patch dividing based feature extraction, which extracts features in a way that preserves the spatial layout of input images; (3) pose-wise classifier sharing, which learns robust classifiers and makes robust predictions; and (4) graphical model based inference, which utilizes the interdependence between different subcategories to update raw estimations to better ones.
