Fuzzy Latent-Dynamic Conditional Neural Fields for Gesture Recognition in Video

With the explosion of data on the internet led to the presence of the big data era, so it requires data processing in order to get the useful information. One of the challenges is the gesture recognition the video processing. Therefore, this study proposes Latent-Dynamic Conditional Neural Fields and compares with the other family members of Conditional Random Fields. To improve the accuracy, these methods are combined by using Fuzzy Clustering. From the result, it can be concluded that the performance of Latent-Dynamic Conditional Neural Fields are lower than Conditional Neural Fields but higher than the Conditional Random Fields and Latent-Dynamic Conditional Random Fields. Also, the combination of Latent-Dynamic Conditional Neural Fields and Fuzzy C-Means Clustering has the highest. This evaluation is tested in a temporal dataset of gesture phase segmentation