Participants: 30 people Duration: 1 hour This seminar provides a foundational overview of mobile robotics and ROS (Robot Operating System) within the scope of the Smart Droplets project. Eurecat explains the technical challenges of autonomous navigation, including sensor noise and […]
Duration: 2 hours Participants: 30 people This seminar explores advanced techniques that enhance computer vision performance in agricultural settings. Starting with a quick review of CNNs, evaluation metrics, and image representations, it introduces state-of-the-art methods such as transformer backbones, multi-scale […]
Duration: 1 hours Participants: 10 people This seminar introduces the principles and tools behind effective image annotation for object detection. It covers best practices for defining classes and regions (bounding boxes, polygons, masks), avoiding data leakage, and ensuring annotation quality […]
Lecturer: Borja Espejo-Garcia Duration: 2 hours Participants: 10 people This workshop introduces key object detection models used in agricultural applications, focusing on Faster R-CNN and YOLO. Participants explore the differences between two-stage and single-stage detectors, comparing trade-offs in accuracy and […]
Lecturer: Borja Espejo-Garcia Duration: 2 hours Participants: 10 people This seminar explores how transfer learning can accelerate weed identification in agriculture using limited training data. Instead of training models from scratch, participants learn to adapt pre-trained vision models (CNNs and […]
Duration: 1 hour Participants: 15 people This lecture introduces Reinforcement Learning (RL) in the context of Agent-Based Modeling (ABM). It explores how RL can be used to train intelligent agents that operate within an ABM, making them adaptive and decision-driven […]
Duration: 1 hour Participants: 30 people This recorded seminar delves into the application of crop hybrid modeling in agriculture. It delved into the pros and cons of each approach, highlighting the potential of combining them to address complex agricultural challenges. […]
Duration: 1 hour Participants: 15 people This presentation delves into the application of Digital Twin technology and AI in Smart Droplets, specifically focusing on WP3. The presentation highlights three key components: a robust data management platform, a powerful digital twin […]
Duration: 2 hours Participants: 15 people The lecture provides an overview of policy approximation in Reinforcement Learning (RL), with a particular emphasis on the use of neural networks. It introduces the concept of policy networks, which learn to map states […]
Lecturer: Borja Espejo-Garcia Duration: 2 hours Participants: 15 people In this event, we analyzed the difference between CNNs and transformers by implementing computer vision systems using both technologies. CNNs are particularly suited for processing grid-like data, such as images, where […]
