Video Understanding: A Predictive Analytics Perspective
| dc.contributor.advisor | Wildes, Richard P. | |
| dc.contributor.author | Zhao, He | |
| dc.date.accessioned | 2022-12-14T16:42:33Z | |
| dc.date.available | 2022-12-14T16:42:33Z | |
| dc.date.copyright | 2022-09-14 | |
| dc.date.issued | 2022-12-14 | |
| dc.date.updated | 2022-12-14T16:42:32Z | |
| dc.degree.discipline | Computer Science | |
| dc.degree.level | Doctoral | |
| dc.degree.name | PhD - Doctor of Philosophy | |
| dc.description.abstract | This dissertation includes a detailed study of video predictive understanding, an emerging perspective on video-based computer vision research. This direction explores machine vision techniques to fill in missing spatiotemporal information in videos (e.g., predict the future), which is of great importance for understanding real world dynamics and benefits many applications. We investigate this direction with depth and breadth. Four emerging areas are considered and improved by our efforts: early action recognition, future activity prediction, trajectory prediction and procedure planning. For each, our research presents innovative solutions based on machine learning techniques (deep learning in particular) and meanwhile pays special attention to their interpretability, multi-modality and efficiency, which we consider as critical for next-generation Artificial Intelligence (AI). Finally, we conclude this dissertation by discussing current shortcomings as well as future directions. | |
| dc.identifier.uri | http://hdl.handle.net/10315/40779 | |
| dc.language | en | |
| dc.rights | Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests. | |
| dc.subject | Computer science | |
| dc.subject.keywords | Computer vision | |
| dc.subject.keywords | Video understanding | |
| dc.title | Video Understanding: A Predictive Analytics Perspective | |
| dc.type | Electronic Thesis or Dissertation |
Files
Original bundle
1 - 1 of 1