In our continuous efforts under the DEFEAT project's Work Package 3 (WP3), we have taken significant strides in enhancing the efficiency and accuracy of sorting operations in construction demolition waste (CDW). With the power of computer vision, our team has successfully developed an advanced system capable of object detection within CDW. This technology aims to revolutionize the way we approach waste management in construction, contributing to more sustainable and environmentally friendly practices.
As part of our commitment to scientific progress and community engagement, the DEFEAT team is thrilled to unveil two interactive web applications. These tools embody the team’s dedication to open access practices and the sharing of knowledge within the scientific community and beyond.
- CDW Annotation Viewer: https://coddannotator.streamlit.app/
This first application is designed for an in-depth exploration of our carefully curated datasets (https://data.mendeley.com/datasets/wds85kt64j/2) . It offers a user-friendly platform to visualize detailed annotations in the collected images of construction waste, providing a unique window into the intricate world of CDW, showcasing the meticulous work involved in dataset creation and annotation.
- CDW Model Inference Demonstrator: https://yolov8inferencetool.streamlit.app/
The second application is an interactive showcase of the cutting-edge object detection models developed in the frame of WP3. Here, users can witness the power of these models in action, observing how they identify and classify various types of waste materials. Beyond demonstrating the technological prowess of the computer vision system, this application also serves as an educational platform and a communication tool for those keen on understanding the practical applications of AI in waste management.