International. Hikvision was ranked #1 in the Scene Classification category at the 2016 ImageNet Large Scale Visual Recognition Challenge.
ImageNet, the global image database resource, has hosted the Large Scale Visual Recognition Challenge (ILSVRC) to create a benchmark within the classification and detection of the category of objects through hundreds of categories, in addition to millions of images. Launched annually since 2010, the challenge attracts participants from more than 50 leading research institutions worldwide.
This ranking is organized through Stanford University, Carnegie Mellon University, the University of Michigan and UNC Chapel Hill. And a high-level impetus from ILSVRC is what will allow researchers to buy on the progress in detection by means of a wide variety of objects. Another driver is the evaluation of the progress of computer vision for large-scale image indexing, allowing for effective removal and annotation.
#1 in scene ranking
Integrating object detection, object localization, video object detection, scene classification and scene syntactic analysis, the five ILSVRC categories have been considered and evaluated against a particular challenge.
For the afternoon of scene sorting, in which Hikvision achieved the No. 1 position, the Hikvision Research Institute used the creation style networks and the not-so-deep residual networks that achieve a better result in much less training time, according to Hikvision's experiments and several improvements made to carry out the training and testing.
"The technical data resulting from the competition can be applied to vehicle detection, license plate recognition, vehicle sub-brand recognition, human detection, human property analysis, facial recognition, image search and much more, greatly improving product performance and application results," said Shiliang Pu, executive vice president of the Hikvision Research Institute. "In the future, the levels of intelligence and automation of machines are going to improve significantly, being used in the sectors of intelligent surveillance, driving assistance systems, intelligent traffic sensor, robotics and unmanned aerial vehicles, just to name a few."
"Since its inception, the Hikvision Research Institute has accumulated an outstanding technical track record," said Cynthia Ho, vice president of Hikvision. "The LSVRC results underscore that the Hikvision Research Institute is at the forefront of computer vision research, and its research results will provide outstanding and sustained technical support for the development of Hikvision's ongoing technology."
