"Scene Parsing, Ego-Positioning and Related Topics"

Past Event Date: 

Meeting room: 

Room 204 - Main Conference Room

Event Type: 

Presenter: Ming-Hsuan Yang, Associate Professor of

Electrical Engineering and Computer Science at

University of California, Merced.

Host: James Coughlan, Ph.D.

 

I will present some of our recent results on how to parse images with labels to better understand the 2D scenes. In

addition, I will present a vision-based approach for ego-positioning in the 3D world based on monocular images.

When time allows I will give a brief overview of other ongoing research on object detection, tracking, segmentation

and recognition in my lab.

.