Deep learning

Researchers develop a new image-processing algorithm that sharpens hazy images by distinguishing true colours and removing visual noise.

A new image dehazing algorithm sharpens both real-world and computer-generated images with unprecedented speed and accuracy.

By linking past observations with future possibilities, a novel framework could help computers predict human actions more accurately.

A deep learning technique developed by A*STAR researchers uses WiFi data to help robots find their way around.

A hybrid approach combining deep learning with Bayesian inference has enabled more accurate, efficient and automatic crack detection.

Floor mats equipped with deep learning could take us one step closer to truly smart homes.

Leveraging both artificial intelligence and field knowledge may prove to be a superior strategy for predicting machine health.

High dynamic range imaging combined with deep learning approaches could improve the ability of autonomous vehicles to recognize traffic lights.

Using deep-learning techniques to locate potential human activities in videos