Besides the presence of Gradiant in the last ICME, held in July in Barcelona, two more research Works carried out by the Multimodal Information Area within VISAGE and PROSSAE projects have been accepted at international conferences.
Enrique Sánchez Lozano, researcher of this area, participated in the 7th International Symposium on Image and Signal Processing and Analysis (ISPA), held in Dubrovnik, Croatia, from September 4th to 6th, presenting the paper “Parameter Constraints for Accurate Face Alignment in the Simultaneous Inverse Compositional Algorithm”, written jointly with Daniel González Jiménez, Multimodal Information Area coordinator, and José Luis Alba Castro, professor from the Signal Theory Group of the University of Vigo. The paper was included into the special session in advanced biometrics, where some other works about face recognition and fingerprinting could be found. In addition, some other works about new technologies such as medical image, crater detection and gesture recognition were included in the ISPA.
Also, the paper entitled “Single- and Cross- Database Benchmarks for Gender Classification Under Unconstrained Settings”, written by Pablo Dago Casas, Daniel González Jiménez, Long Long Yu and José Luis Alba Castro has been accepted in the 1st IEEE International Workshop on Benchmarking Facial Image Analysis Technologies
(BeFIT), that will be held in November in Barcelona and that is organized in conjunction with the International Conference on Computer Vision (ICCV), one of the most prestigious conferences in computer vision.
This work is focused in gender classification (male/female) from facial features, and attempts to establish a common test framework to allow different authors to compare their results in an objective and reliable way. At the same time, it reflects the efforts made to adequate and truthfully evaluate the demographic classifier in real application environments, a trendy tendency in facial image processing that shows the need to increase the real world applicability of these technologies.
With these publications, Gradiant presents the improvements achieved in facial image processing, and hopes to keep increasing its contributions to these fields.