Metalens taps AI to improve ultra thin imaging

Researchers in Korea have used deep learning compensation to improve the end-to-end imaging using a high volume metalens rather than a traditional lens.

This could open up dramatically thinner optics in smartphones, virtual reality (VR), and augmented reality (AR) devices.

The system developed by the Pohang University of Science and Technology (POSTECH) uses a 10mm diameter ultra-thin metalens composed of tiny nanostructures built with nanoprint lithography.

The system pairs a mass-produced metalens with a specialized image restoration framework driven by deep learning. By combining advanced optical hardware with artificial intelligence (AI), the team has achieved high-resolution, aberration-free, full-colour images, all while maintaining a compact form factor.

• Metalens expands from light to sound
• Metalenz, launch metasurface optics
• ST teams for metalens tech

The metalens itself is fabricated using nanoimprint lithography, a scalable and cost-effective method, followed by atomic layer deposition, allowing for large-scale production of these lenses.

The metalens is designed to focus light efficiently but, like most metalenses, suffers from chromatic aberration and other distortions due to its interaction with light of different wavelengths. To address this, the deep learning model is trained to recognize and correct the colour distortions and blurring caused by the metalens. This approach is unique because it learns from a large dataset of images and applies these corrections to future images captured by the system.

The image restoration framework uses adversarial learning, where two neural networks are trained together. One network generates corrected images, and the other assesses their quality, pushing the system to improve continuously. Additionally, advanced techniques like positional embedding help the model understand how image distortions change depending on the viewing angle. This results in significant improvements in the restored images, particularly in terms of color accuracy and sharpness across the entire field of view.

“This deep-learning-driven system marks a significant advancement in the field of optics, offering a new pathway to creating smaller, more efficient imaging systems without sacrificing quality,” said Prof Junsuk Rho at POSTECH

The ability to mass-produce high-performance metalenses, combined with AI-powered corrections, brings us closer to a future where compact, lightweight, and high-quality imaging systems become the norm in both commercial and industrial applications.