SkySearch | Data Science Capstone Project скачать в хорошем качестве

SkySearch | Data Science Capstone Project

As part of their Capstone Project, a team of MS in Applied Data Science (MS-ADS) students designed and built SkySearch, an unmanned aerial vehicle (UAV) vision system that tests a UAV's ability to locate and navigate to almost any object. Traditional object detection algorithms such as YOLO have been widely adopted by UAV operators, but they are inherently limited by their pre-defined object categories, which only include a small subset of objects. SkySearch addresses this limitation by integrating zero-shot learning and multimodal large language models, allowing the system to identify and locate objects beyond a predefined set of labels. This capstone project illustrates how MS-ADS students apply advanced techniques in computer vision, machine learning, and large language models to real-world challenges.