In this paper, we present a statistical characterization of tile decoding time of 360° videos encoded via HEVC that considers different tiling patterns and quality levels (i.e., bitrates). In particular, we present results for probability density function estimation of tile decoding time based on a series of experiments carried out over a set of 360° videos with different spatial and temporal characteristics. Additionally, we investigate the extent to which tile decoding time is correlated with tile bitrate (at chunk level), so that DASH-based video streaming can make possible use of such an information to infer tile decoding time. The results of this work may help in the design of queueing or control theory-based adaptive bitrate (ABR) algorithms for 360° video streaming.
Developing machine learning models for image classification problems involves various tasks such as model selection, layer design, and hyperparameter tuning for improving the model performance. However, regarding deep learning models, insufficient model interpretability renders it infeasible to understand how they make predictions. To facilitate model interpretation, performance analysis at the class and instance levels with model visualization is essential. We herein present an interactive visual analytics system to provide a wide range of performance evaluations of different machine learning models for image classification. The proposed system aims to overcome challenges by providing visual performance analysis at different levels and visualizing misclassification instances. The system which comprises five views - ranking, projection, matrix, and instance list views, enables the comparison and analysis different models through user interaction. Several use cases of the proposed system are described and the application of the system based on MNIST data is explained. Our demo app is available at https://chanhee13p.github.io/VisMlic/.