What is it about?
This paper explores how AI and low-cost tools can identify and track fabrics to reduce textile waste. By using fabric images and special markers, it improves recycling and traceability, helping create a sustainable circular economy for clothing.
Featured Image
Photo by Greg Rosenke on Unsplash
Why is it important?
The textile industry produces massive waste and pollution, recycling less than 1% of garments globally. Identifying and tracing fabrics accurately is crucial for enabling recycling, reducing landfills, and supporting sustainability laws. By making these processes low-cost and scalable, this solution can help protect the environment and promote a circular economy for textiles.
Perspectives
I started this work because I love clothes but then discovered the damage that my favorite fast fashion clothing was doing to the environment and to human health. How many of us ask - "Where do my clothes come from? When I am done with them, where do they go and what happens to them?". Textile waste is a large contributor to planetary pollution. Methods like those described in this paper can help us create a sustainable circular economy to protect our world from the scourge of textile pollution.
Danika Gupta
The Harker School
Read the Original
This page is a summary of: Towards Improved Sustainability in The Textile Lifecycle with Deep Learning, July 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3674829.3675077.
You can read the full text:
Resources
Object Detection of Fabric Codes - Dataset 1
A bounding box labeled dataset of fabric codes. Suitable for YOLOv8 training.
Fabric code object detection - Dataset 2
Bounding box labeled dataset for object detection of fabric codes. Unlike dataset 1 - these images are of washed fabrics where the wash has worn off some of the code.
Contributors
The following have contributed to this page
