Skip to the content.

Deep Learning in Science and Engineering

Zhangyang (Atlas) Wang, Department of Computer Science & Engineering
A (Fairly Rough) Tour of Our Recent Works in Computer Vision, Machine Learning, and Their Applications

Abstract

In this talk, I will give a quick and rough overview of various research activities conducted by the Visual Informatics@Texas A&M (VITA) group (http://www.atlaswang.com/group.html). The research of VITA is centered at the techniques of deep learning, as well as the data modality of image/video (defined in its broadest sense). We are fascinated by both theories and applications. I will categorize and introduce our recent research into four areas:

  1. low-level computer vision (image restoration and enhancement);
  2. high-level computer vision (recognition, segmentation, detection/tracking, re-identification);
  3. core machine learning model & theory (deep network training, sparse optimization, learning under data/resource budgets);
  4. some medical and healthcare applications (predicting emotional disorder, and Alzheimer’s disease).

Given the talk's overview nature, follow-up questions are more than welcome from the audience, and I will be happy to go into technical details of any project upon request.