VIEW-INVARIANT ACTION RECOGNITION IN DYNAMIC SCENES VIA SIM2REAL TRANSFER
| dc.contributor.author | Wang, Yuhan | |
| dc.contributor.chair | Ferrari, Silvia | en_US |
| dc.contributor.committeeMember | Hariharan, Bharath | en_US |
| dc.date.accessioned | 2024-04-05T18:36:01Z | |
| dc.date.available | 2024-04-05T18:36:01Z | |
| dc.date.issued | 2023-08 | |
| dc.description | 62 pages | en_US |
| dc.description.abstract | Learning view-invariant representation is essential to improving feature extraction for action recognition. Existing approaches cannot effectively capture details for human actions due to fast-paced gameplay and implicit view- dependent representation. In this paper, it goes beyond recognizing human actions from a fixed view and focusing on action recognition from an arbitrary view. This paper purposes a method to build an efficient data generating pipeline due to lack of original input data. This paper also provides a pipeline combining capturing modified I3D human actions features and use Multilayer Perception to achieve human action recognition and classification. The use of information captured from combination of virtual and real-life data, as well as different viewing angles, leads to high classification performance. | en_US |
| dc.identifier.doi | https://doi.org/10.7298/h4vt-nv30 | |
| dc.identifier.other | Wang_cornell_0058O_11820 | |
| dc.identifier.other | http://dissertations.umi.com/cornell:11820 | |
| dc.identifier.other | Wang_cornell_0058O_11839 | |
| dc.identifier.other | http://dissertations.umi.com/cornell:11839 | |
| dc.identifier.uri | https://hdl.handle.net/1813/114413 | |
| dc.language.iso | en | |
| dc.subject | Computer vision | en_US |
| dc.subject | Machine Learning | en_US |
| dc.subject | MLP | en_US |
| dc.title | VIEW-INVARIANT ACTION RECOGNITION IN DYNAMIC SCENES VIA SIM2REAL TRANSFER | en_US |
| dc.type | dissertation or thesis | en_US |
| dcterms.license | https://hdl.handle.net/1813/59810.2 | |
| thesis.degree.discipline | Mechanical Engineering | |
| thesis.degree.grantor | Cornell University | |
| thesis.degree.level | Master of Science | |
| thesis.degree.name | M.S., Mechanical Engineering |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- Wang_cornell_0058O_11820.pdf
- Size:
- 4.37 MB
- Format:
- Adobe Portable Document Format