Our Gauge Equivariant CNNs help machines see and understand like humans to improve AI experiences

At Qualcomm AI Research, we’re on a mission to help machines see and understand like humans, and we’re one step closer. Unlike traditional Convolutional Neural Networks (CNN), our Gauge Equivariant CNNs (G-CNNs) can analyze image data on any curved space or geometry, on-device. This means highly efficient AR, VR, autonomous car, and drone experiences. Hear a high-level explanation is this animation video, narrated by Professor Max Welling, Vice President, Technology at Qualcomm Technologies Netherlands B.V.

Read more: https://www.qualcomm.com/news/onq/2019/05/02/helping-machines-see-and-understand-humans
Learn more: https://arxiv.org/abs/1902.04615

Видео Our Gauge Equivariant CNNs help machines see and understand like humans to improve AI experiences канала Qualcomm