Human motion recognition in video data has several interesting applications in fields such as gaming, senior/assisted-living environments, and surveillance. In these scenarios, we may have to consider adding new motion classes (i.e., new types of human motions to be recognized), as well as new training data (e.g., for handling different type of subjects). Hence, both the accuracy of classification and training time for the machine learning algorithms become important performance parameters in these cases. In this article, we propose a knowledge-based hybrid (KBH) method that can compute the probabilities for hidden Markov models (HMMs) associated with different human motion classes. This computation is facilitated by appropriately mixing features from two different media types (3D motion capture and 2D video). We conducted a variety of experiments comparing the proposed KBH for HMMs and the traditional Baum-Welch algorithms. With the advantage of computing the HMM parameter in a noniterative manner, the KBH method outperforms the Baum-Welch algorithm both in terms of accuracy as well as in reduced training time. Moreover, we show in additional experiments that the KBH method also outperforms the linear support vector machine (SVM). ACM Reference Format:Suk, M., Ramadass, A., Jin, Y., and Prabhakaran, B. 2012. Video human motion recognition using a knowledge-based hybrid method based on a hidden Markov model.

This paper focuses on the issue of improving the quality of low level 2D feature extraction for human action recognition. For instance, existing algorithms such as the Optical Flow algorithm detects noisy and irrelevant features because of its lack of ground truth data sets for complex scenes. For these features, it is difficult to extract data such as coordinate positions of the features, velocity and the direction of the moving objects, and the differential data information between different frames. Extracting such low level feature data is one of the major steps involved in video based Human action recognition. The paper proposes an extended Optical Flow algorithm focusing on human actions. This uses a Frame Jump technique along with thresholding of unwanted features to overcome the problems due to complex scenes. Frame Jump restricts to detecting only useful features by removing other features detected by the existing Optical Flow algorithm. In addition to the above, it also elucidates the integration of the proposed technique with other feature extraction algorithms.

No abstract