HIGHLIGHTS
- who: Harideep Nair from the Adobe Research, United States University of Technology Sydney have published the Article: SemNet: Learning semantic attributes for human activity recognition with deep belief networks, in the Journal: (JOURNAL)
- what: The authors expand the RBM into a hierarchical representation, wherein relevant semantic concepts are revealed at the higher levels. The authors demonstrate that the authors can find semantic concepts similar to attributes like "arm up" and "arm down," even though no information with regards to these attributes was given during the training process. This approach has also proven effective in computer . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.