Last month Google released its next generation of on-device computer vision networks – MobileNewtV2. Google is claiming a substantially faster process for the same accuracy across the entire latency spectrum. HPC Wire brought this news to our attention in their article, “Google Releases Improved Neural Networks for Vision Recognition.”

Various Google-developed machine learning and deep learning tools have become widely adopted across many domains. MobileNetV2 advances mobile visual recognition including classification, object detection and semantic segmentation.

MobileNets are small, low-latency, low-power models with the parameters to meet the resource constraints of a variety of use cases. MobileNetV2 improves the state-of-the-art performance of mobile models on multiple tasks and benchmarks as well as across a spectrum of different model sizes.

The new version of MobileNet has several properties that make it suitable for mobile applications and allows very memory-efficient inference and utilizes standard operations present in all neural frameworks.

Melody K. Smith

Sponsored by Access Innovations, the world leader in taxonomies, metadata, and semantic enrichment to make your content findable.