In this paper we present a novel two stage algorithm for improving video coding efficiency. The proposed method combines video cut detection and adaptive GOP structure. At first, we have proposed a new technique of frames' comparison for the shot cut detection. The majority of existing methods compare pairs of successive frames. We compare actual frame with its motion estimated prediction. We also present adaptive threshold. The efficiency of novel technique for video cut detection was confirmed through experiment and compared to the commonly used ones in the terms of recall and precision. The next step is to situate I frames to the positions of detected cuts during the process of video encoding. Finally the proposed method is verified by simulations and the obtained results are compared with fixed GOP structures of sizes 4, 8, 12, 16, 32, 64, 128 and GOP structure with length of entire video. Proposed method achieved the gain in bit rate from 15,33% to 50,59%, while not degrading PSNR in comparison to simulated fixed GOP structures.

A new detection method for cognitive impairments is presented utilizing an eye tracking signals in a text reading test. This research enhances published articles that extract combination of various features. It does so by processing entire eye-tracking records either in time or frequency whereas applying only basic signal pre-processing. Such signals were classified as a whole by Convolutional Neural Networks (CNN) that hierarchically extract substantial features scatter either in time or frequency and nonlinearly binds them using machine learning to minimize a detection error. In the experiments we used a 100 fold cross validation and a dataset containing signals of 185 subjects (88 subjects with low risk and 97 subjects with high risk of dyslexia). In a series of experiments it was found that magnitude spectrum based representation of time interpolated eye-tracking signals recorded the best results, i.e. an average accuracy of 96.6% was reached in comparison to 95.6% that is the best published result on the same database. These findings suggest that a holistic approach involving small but complex enough CNNs applied to properly pre-process and expressed signals provides even better results than a combination of meticulously selected well-known features.

No abstract

No abstract

No abstract