STUDENT ATTENDANCE SYSTEM IN CROWDED CLASSROOMS: We release a realistic full-annotated dataset of images of a classroom with around 70 students in 25 sessions, taken during 15 weeks. Ten face recognition algorithms based on learned and handcrafted features are evaluated using a protocol that takes into account the number of face images per subject used in the gallery. In our experiments, the best one has been FaceNet, a method based on deep learning features, achieving around 95% of accuracy with only one enrollment image per subject.
Mery, D.; Mackenney, I.; and Villalobos, E. Student Attendance System in Crowded Classrooms using a Smartphone Camera. In 2019 IEEE Winter Conference on Applications of Computer Vision (WACV2017), 2019. [ PDF ]
- Simple implementation: See GitHub repository.
- Full implementation and dataset: See GitHub repository.
RECOGNITION OF FACIAL ATTRIBUTES AND FACE RECOGNITION: We developed a computer vision algorithm to recognize automatically facial attributes (such as gender, expressions, race, etc.) and faces. Our approach is able to deal with a larger degree of variability in ambient lighting, pose, expression, occlusion, face size and distance from the camera.
Mery, D.; Bowyer, K. (2015): Automatic facial attribute analysis via adaptive sparse representation of random patches. Pattern Recognition Letters (accepted May 2015). [ PDF ]
DETECTION IN COMPLEX OBJECTS USING X-RAY TESTING. We developed a system that is able to detect parts of interest inside of a complex object using multiple X-ray views. In this video we show how to detect all objects of this pen case in only 6 seconds.
Mery, D. (2015): Inspection of Complex Objects Using Multiple X-ray Views. IEEE Transactions on Mechatronics, 20(1):338-347. [ PDF ]
AUTOMATED DETECTION OF FISHBONES IN SALMON FILETES: X-ray testing is playing an increasingly important role in quality assurance of foods. We developed an X-ray machine vision approach to detect fishbones in fishfillets automatically. In the experiments, we used a high resolution flat panel detector that is able to take up to 6 millions pixels digital X-ray image. In only two seconds we are able to detect 95% of the fishbones automatically using 24 features and support vector machines. We believe that the proposed approach opens new possibilities in the field of automated visual inspection of fish.
Mery, D.; Lillo, I.; Loebel, H.; Riffo, V.; Soto, A.; Cipriano, A.; Aguilera, J.M. (2011): Automated Fish Bone Detection using X-ray Testing. Journal of Food Engineering, 105(2011):485-492. [ PDF ]
AUTOMATED VISUAL INSPECTION OF TORTILLAS: We developed a computer vision method to classify the quality of corn tortillas automatically according to five hedonic sub-classes given by a sensorial panel. The proposed methodology analyzed 750 corn tortillas obtained form 15 different Mexican small, medium and large commercial stores. Using only 64 features and support vector machines we are able to detect 95% of all classes. We believe that the proposed methodology opens up new possibilities in the field of automated visual inspection of tortillas.
Mery, D.; Chanona-Perez, J.; Soto, A.; Aguilera, J.M.; Cipriano, A.; Velez-Riverab, N; Arzate-Vazquez, I, Gutierrez–Lopez, G. (2010): Quality Classiﬁcation of Corn Tortillas using Computer Vision. Journal of Food Engineering 101(4):357-364. [ PDF ]