Complexity-Aware Assignment of Latent Values in Discriminative Models for Accurate Gesture Recognition
This work addresses gesture recognition from video streams, offering an incremental improvement for applications in human-computer interaction.
The paper tackles the problem of improving gesture recognition accuracy by proposing a novel methodology to assign latent values in discriminative models based on gesture complexity, resulting in up to a 10% increase in accuracy compared to arbitrary assignments.
Many of the state-of-the-art algorithms for gesture recognition are based on Conditional Random Fields (CRFs). Successful approaches, such as the Latent-Dynamic CRFs, extend the CRF by incorporating latent variables, whose values are mapped to the values of the labels. In this paper we propose a novel methodology to set the latent values according to the gesture complexity. We use an heuristic that iterates through the samples associated with each label value, stimating their complexity. We then use it to assign the latent values to the label values. We evaluate our method on the task of recognizing human gestures from video streams. The experiments were performed in binary datasets, generated by grouping different labels. Our results demonstrate that our approach outperforms the arbitrary one in many cases, increasing the accuracy by up to 10%.