Team
Lucas, Fred, Sayeed, Chandana, Jaspreet
Categories
Artificial Intelligence, Data Analysis, Deep Learning, Machine Learning, Python
Overview
Our project focuses on enhancing robotic perception by developing a deep learning model capable of real-time object detection within its environment. Beyond traditional detection, the model also generates rich image and object-level embeddings, or tokens, which subsequent models can leverage to support autonomous driving tasks. By combining accurate object recognition with the creation of structured, informative representations, our system enables a robot to understand its surroundings and provide meaningful data for higher-level decision-making processes, all optimized for real-time performance critical to autonomous navigation.
Real Time Object Detection
Real-time perception system for autonomous driving: Our deep learning model detects and identifies objects such as vehicles, pedestrians, and street infrastructure, while generating embeddings to support downstream driving tasks.
