Welcome to Ear Recognition Laboratory Homepage at University of Science & Technology Beijing (USTB).
The website was set up formally in January, 2005. We hope to introduce our research results on ear recognition and multimodal recognition fusing face and ear, and also provide an online platform for the interrelated academic exchanges.
As one of the first academic teams which study ear recognition in China, we carried out the research in 2001. The principal of the Ear Recognition Laboratory are Prof. Zhichun Mu and A/Prof. Li Yuan who are with School of Automation and Electrical Engineering at USTB. Under the support in financing by the National Natural Science Foundation of China, we have acquired some delightful production. Based on the progress made on the ear identification project, and from the view point to achieve non-cooperative identification, we extend our research interests to the multi-modal identification technology, which is based on the ear and face information integration and 3D ear recognition. We have established the USTB ear image database and it is available for science research.
Ear Identification is a newly emerged biometric technology. It now becomes one of the most popular studies in the field of biometric. The ear recognition website supports the research in the areas of pattern recognition, image processing and computer vision. Our latest research results and the USTB image database are also available. Any comments and suggestions on our research work from you are welcome!
In 2016, we cooperated with Helloear Co. Ltd. and established a database named ‘USTB-Helloear’, which photographed under uncontrolled conditions. The USTB-Helloear database is a large profile database that contains more than 610,000 images from 1570 subjects taken from both the left and right side. The main distinguishing feature of the images in this database is that they were taken under uncontrolled conditions with illumination and pose variation. In addition, all of individuals were required to not particularly care about ear occlusions. Therefore, 30% of objects had the control groups with different level of ear occlusions. The database will be provided free of charge to download recently.