2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2015)
Boston, MA, USA
June 7, 2015 to June 12, 2015
Sameh Khamis , University of Maryland, College Park, 20742, United States
Larry S. Davis , University of Maryland, College Park, 20742, United States
Action recognition is a fundamental problem in computer vision. However, all the current approaches pose the problem in a multi-class setting, where each actor is modeled as performing a single action at a time. In this work we pose the action recognition as a multi-label problem, i.e., an actor can be performing any plausible subset of actions. Determining which subsets of labels can co-occur is typically treated as a separate problem, typically modeled sparsely or fixed apriori to label correlation coefficients. In contrast, we formulate multi-label training and label correlation estimation as a joint max-margin bilinear classification problem. Our joint approach effectively trains discriminative bilinear classifiers that leverage label correlations. To evaluate our approach we relabeled the UCLA Courtyard dataset for the multi-label setting. We demonstrate that our joint model outperforms baselines on the same task and report state-of-the-art per-label accuracies on the dataset.
Legged locomotion, Yttrium, Optimization, Prediction algorithms, Algorithm design and analysis, Silicon
S. Khamis and L. S. Davis, "Walking and talking: A bilinear approach to multi-label action recognition," 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Boston, MA, USA, 2015, pp. 1-8.