Scientific work

At Visage Technologies, we believe that continuous learning is the key to success. We love staying up-to-date with the latest trends through internal workshops, summer schools, industry-leading conferences, our own library… but also giving back to the community by organizing events such as our Computer Vision Talks conference.

We take pride in the fact that 70% of our team is fully dedicated to research and development. This page provides an overview of our team members’ scientific work that has been published so far.

Published papers

• Bešenić, K., Ahlberg, J. & Pandžić, I.S. Picking out the bad apples: unsupervised biometric data filtering for refined age estimation. Vis Comput (2022).

• Gogić, Ivan, Martina Manhart, Igor S. Pandžić, and Jörgen Ahlberg. Fast facial expression recognition using local binary features and shallow neural networks. The Visual Computer 36, no. 1 (2020): 97-112.

• Markuš, Nenad, Ivan Gogić, Igor Pandžić, and Jörgen Ahlberg. Memory-efficient Global Refinement of Decision-Tree Ensembles and its Application to Face Alignment. In British Machine Vision Conference (BMVC), Northumbria University, Newcastle upon Tyne, UK, 3-6 September 2018, pp. 1-11. The British Machine Vision Association and Society for Pattern Recognition, 2018.

• Gogić, Ivan, Jörgen Ahlberg, and Igor S. Pandžić. Regression-based methods for face alignment: A survey. Signal Processing (2020): 107755.

• Dijana Vitas; Martina Tomic; Matko Burul. Traffic Light Detection in Autonomous Driving Systems. IEEE Consumer Electronics Magazine, Vol. 9, Iss. 4, pp. 90 – 96, 2020, DOI: 10.1109/mce.2020.2969156

• Dijana Vitas; Martina Tomic; Matko Burul. Image Augmentation Techniques for Cascade Model Training. 2018 Zooming Innovation in Consumer Technologies Conference (ZINC), 30-31 May 2018, DOI: 10.1109/ZINC.2018.8448407

• Bešenic, Krešimir, Jörgen Ahlberg, and Igor S. Pandzic. Unsupervised Facial Biometric Data Filtering for Age and Gender Estimation. (2019).

• N. Markuš, I. S. Pandžić, and J. Ahlberg. Learning Local Descriptors by Optimizing the Keypoint-Correspondence Criterion: Applications to Face Matching, Learning from Unlabeled Videos and 3D-Shape Retrieval. IEEE Transactions on Image Processing, Vol. 28, Issue 1, pp. 279-290, January 2019.

• N. Markuš, M. Frljak, I. S. Pandžić, J. Ahlberg, and R. Forchheimer. Eye pupil localization with an ensemble randomized trees. Pattern Recognition, Vol. 47, Issue 2, pp. 578-587, February 2014.

• N. Markuš, I. S. Pandžić, and J. Ahlberg. Learning Local Descriptors by Optimizing the Keypoint-Correspondence Criterion. International Conference on Pattern Recognition (ICPR), Cancun, Mexico, December 2016, pp. 2380-2385.

• N. Markuš, M. Frljak, I. S. Pandžić, J. Ahlberg, and R. Forchheimer. High-performance face tracking. ACM 3rd International Symposium on Facial Analysis and Animation (FAA), Vienna, Austria, September 2012.

• F. Dornaika and J. Ahlberg. Fitting 3D Face Models for Tracking and Active Appearance Model Training. Image and Vision Computing, Vol. 24, pp. 1010–1024, 2006.