MIT researchers enhance AI training with synthetic images
The new technology can help to address the challenges of data acquisition in machine learning and better train models (Image: MIT News)
MIT researchers have been working to reshape AI training using a system that leverages text-to-image models to generate diverse synthetic images for machine learning
Register for FREE to keep reading
Join 10,000+ vision professionals driving innovation in automation, AI and imaging with:
- Expert insights on vision, robotics, AI & embedded tech
- Newsletters and features covering the full imaging landscape
- Visionaries series: leadership strategies in imaging
- Free panels on smart manufacturing & autonomy
- White Papers & updates for smarter integration
Sign up now
Already a member? Log in here
Your data is protected under our privacy policy.
