Real time face detection combined with eyegaze tracking can provide a means of user input into gaming environment. Game and CE system designers use facial eye-gaze information in various ways to enhance UI design providing smarter modes gameplay interaction modalities that are sensitive users behaviors mood. Here we review earlier approaches, using wearable sensors, or enhanced IR illumination. Our technique only requires video feed from low-resolution user-facing camera. The algorithm is...

Due to the complexity and size of modern software systems, amount logs generated is tremendous. Hence, it infeasible manually investigate these data in a reasonable time, thereby requiring automating log analysis derive insights about functioning systems. Motivated by an industry use-case, we zoom-in on one integral part automated analysis, parsing, which prerequisite deriving any from logs. Our investigation reveals problematic aspects within parsing field, particularly its inefficiency...

Real-time face detection combined with eye-gaze tracking and analysis can provide a new means of user input in gaming environment. Game designers use facial information various ways designing the interface (UI) to smarter modes interaction players.

Large language models have gained widespread popularity for their ability to process natural inputs and generate insights derived from training data, nearing the qualities of true artificial intelligence. This advancement has prompted enterprises worldwide integrate LLMs into services. So far, this effort is dominated by commercial cloud-based solutions like OpenAI's ChatGPT Microsoft Azure. As technology matures, however, there a strong incentive independence major cloud providers through...

This paper presents the combined effect of two most challenging problems low light: auto-focus (AF) error and how it impacts on automatic red-eye removal algorithms. We propose an algorithm to overcome issues which consists understanding level blur in input image simulation Bokeh corrected pixels. The is designed avoid complex computations cannot run embedded devices such as digital cameras.

Due to the complexity and size of modern software systems, amount logs generated is tremendous. Hence, it infeasible manually investigate these data in a reasonable time, thereby requiring automating log analysis derive insights about functioning systems. Motivated by an industry use-case, we zoom-in on one integral part automated analysis, parsing, which prerequisite deriving any from logs. Our investigation reveals problematic aspects within parsing field, particularly its inefficiency...

With the tremendous log volume generated by modern systems, automated analysis becomes indispensable for discovering crucial insights into behavior of running systems. The foremost step subsumed in an pipeline, called parsing, significantly influences its entire performance. Despite significance, parsing lacks quality implementations and, practice, suffers from fundamental limitations, thereby creating a bottleneck valuable at line level. As consequence, my PhD thesis, I endeavor to explore...

Machine learning is becoming a key technology to make systems smarter and more powerful. Unfortunately, training large capable ML models resource-intensive requires high operational skills. Serverless computing an emerging paradigm for structuring applications benefit from on-demand resources achieve horizontal scalability while making easier consume. As such, it ideal substrate the often ad-hoc task of deep has strong potential democratize access techniques. However, design serverless...