报告题目：Feature Image Pyramid Network for Object Detection in Remote Sensing Imagery
报告人: Pourya Shamsolmoali 博士
主持人：吕岳 教授
报告时间：10月9日15:30 pm
报告地点：信息楼341
报告人简介:
Pourya Shamsolmoali is a Postdoctoral Researcher at the Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong University. He is also a Research Fellow at the Institute of Computer Vision, ETS, Montreal, Canada. In 2020, he was selected as a talented researcher by Shanghai Jiao Tong University. His research focuses on multi-task learning, domain adaptation, and deep learning theory. Pourya Shamsolmoali as the first and corresponding author published more than 20 SCI papers in renowned journals such as IEEE TGRS, IEEE JSTAE, Information Fusion, TOMM, and Neurocomputing. His paper awarded as the best paper in IET Image Processing and 2 of his papers were selected as highly cited papers in PRL.
报告摘要:
Detection of objects is extremely important in various aerial vision-based applications. Over the last few years, the methods based on convolution neural networks have made substantial progress. However, because of the large variety of object scales, densities, and arbitrary orientations, the current detectors struggle with the extraction of semantically strong features for small-scale objects by a predefined convolution kernel. In this talk, I will cover all these topics and propose our rotation equivariant feature image pyramid network for object detection in optical remote sensing images.
