Vision Transformers can outperform other methods like RNNs in classification tasks in the medical domain. MedMNIST (v2) is a collection of biomedical images, and contains eight 2D image datasets for a multi-class classification task. Using a pre-trained Vision Transformer model, and fine-tuning it for each dataset, we were able to outperform almost all of the … Read more
How well can we detect stress from sensor data from wearables? And how to develop systems for it fast? You can find the full project and code on GitHub and a paper here (PDF). Or just visit the interactive live demo showcasing the whole project right now!
Can you predict if a self-described pornography-addict will relapse – based on his previous mental condition? We used machine learning to answer this question based on data collected by polish collaborators. Users used a mobile app for tracking their addictions and mental state. You can find a full project description and code on GitHub.
Imagine getting recommendations, for example, for music or TV shows, based on similar people you pass by on the street. No server backend needed, all via device-to-device communication. We designed and implemented such a system. You can find our full article published in IEEE Access here (PDF) and the code on GitHub. We build the … Read more
We released AnaLoc on GitHub. AnaLoc is a microservice-based, scalable architecture for stream and batch processing of industrial sensor data. The code on GitHub shows the analysis of location data of autonomous vehicles in the modern factory. The architecture is built on Spring Boot and Apache Flink and is deployed via Docker Compose. More details … Read more
