Abstract: There are many benchmark data sets for video action recognition, such as UCF-101, Activity Net and Kinetics data set. They are all the label scheme based on image classification, ...
Abstract: Conventional domain adaptation methods usually resort to deep neural networks or subspace learning to find invariant representations across domains. However, most deep learning methods ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback