Rank One announces a major enhancement to the ROC face recognition SDK with the newly released version 1.20. The most significant improvement is in face recognition accuracy, with as much as a 40% reduction in error rates.
The embedded algorithm algorithm that ships with the ROC SDK, which is designed for resource-constrained applications such as mobile devices or high-throughput video, did not have a change in accuracy, but did receive a 10% improvement in speed.
Other improvements delivered in v1.20 include an overhaul of the face detection algorithm resulting in both accuracy and speed improvements, the “Ethnicity” classifier has been changed to a “Geographic Origin” classifier, profile faces (i.e., faces whose Yaw pose angle is greater than 60 degrees) can now be manually enrolled, a new classifier for detecting facial drawings was added, the (optional) CUDA GPU enrollment pipeline was made more efficient, new licensing mechanisms were added, and new API wrappers for Python 3 and the Lua programming language were added.
The performance of the v1.20 algorithm is published in the most recent NIST FRVT Ongoing report (Nov. 19, 2019) under the submission `rankone-008’, and was one of the only submissions to achieve better than median performance in all measured categories. A detailed description of the improvements delivered in this release can be found in our blog post.