| |
 Multimodal 3D Understanding Of Dynamic Scenes Speaker (s):
Narendra Ahuja
Research Professor,
Departments of Electrical and Computer Engineering, Coordinated Science Laboratory and Beckman Institute, University of Illinois
| Date: Time: Venue: | | 12 August 2024, Monday 10:30am – 11:30am School of Computing &
Information Systems 2 (SCIS 2)
Level 2, Seminar Room 2-7
Singapore Management University
90 Stamford Road
Singapore 178903 Please register by 11 August 2024. We look forward to seeing you at this research seminar. 
|
|
About the Talk This talk presents an overview of some of our recent research projects, pursuing the following capabilities in various combinations:
1. 3D reconstruction of moving articulated objects, such as human and animal activities, from a single short monocular video, with possible text descriptions
2. Hierarchical representation, and simultaneous recognition and segmentation of the activities in multi-activity videos, captured from an uncalibrated camera
3. Joint audio-visual scene understanding, and video guided separation of sounds from multiple sources in mixed audios
4. Methods for explainable, efficient, multilabel and metric learning, for problems in and out of training distribution
Time permitting, we will also review some applications of these capabilities.
About the Speaker Narendra Ahuja (https://migrate2wp.web.illinois.edu/narendra-ahuja/) is a Research Professor and past Donald Biggar Willet Professor at the University of Illinois at Urbana-Champaign in Departments of Electrical and Computer Engineering, Computer Science, and Coordinated Science Laboratory. His research is in computer vision and machine learning, on integrated use of multiple sources of scene information including images, videos, audio and text to extract and represent spatiotemporal structure of scenes, construct three-dimensional and other descriptions, texture modeling, visual motion understanding, image and video synthesis, sensors for computer vision, explainable efficient learning architectures, and applications including visual communication, image manipulation, healthcare, agriculture, railroads and navigation. He has co-authored three books, published about 500 journal and conference papers, and holds four patents. He has graduated 80 graduate and 100 undergraduate research students. His algorithms have been used by companies like General Electric, CMC, Honeywell, Westinghouse, and Advanced Technology, some incorporated into products. He is the founding director of the first of the IIITs in India, and the founding director of the Information Technology Research Academy in India. He has received several awards from and is a fellow of IEEE, ACM, AAAI, IAPR, AAAS and SPIE. |
