Improving the real-world applicability of visual tracking and image classification

Add to your list(s) Download to your calendar using vCal

• Jongwon Choi, Chung-Ang University, Seoul
• Friday 22 January 2021, 10:00-11:00
• https://bham-ac-uk.zoom.us/j/84307498855?pwd=S1NhS2JmVEVrci9wV0U5Qlo0Q3A2dz09.

If you have a question about this talk, please contact Martin Rudorfer.

Even though many AI-based methods have improved recently the performance of various computer vision tasks, the methods should overcome several obstacles for their real-world applicability. In this presentation, my work will be introduced to improve the real-world applicability of AI for the tasks of visual tracking and image classification. With the visual tracking algorithms, the users can track one specific target from the consecutive video frames. AI-based methods dramatically improve the performance of visual trackers, but the slow speed of AI-based trackers prohibits AI-based trackers from being used in the real world. To solve the problems, we propose several methods to reduce the computational costs of visual trackers to let them run in real-time. In addition, we also introduce our research on domain adaptation to improve the performance of image classification across the various test environment. The domain adaptation improves the classification performance even though the label information of the test environment is not given when the training and test environments are different from each other. Thus, through the domain adaptation scheme, we can adapt the pre-trained AI network into the novel environment without any label cost.

This talk is part of the IRLab Seminars: Robotics, Computer Vision & AI series.

This talk is included in these lists:

• IRLab Seminars: Robotics, Computer Vision & AI
• https://bham-ac-uk.zoom.us/j/84307498855?pwd=S1NhS2JmVEVrci9wV0U5Qlo0Q3A2dz09

Note that ex-directory lists are not shown.