Comparison of machine and biological intelligence.
How deep neuronal networks learn image content?
Role of local contrast in convolutional kernels and receptive-fields of neurons.
Education
2014 - 2017
Ph.D. in Computer Science
Universitat Autònoma de Barcelona, Barcelona, Spain
Thesis: "Computational Model of Visual Perception: From Colour to Form" Director: Dr. Alejandro Párraga
2011 - 2013
MSc. in Computer Vision and Robotics (Erasmus Mundus)
Université de Bourgogne, France & Universitat de Girona, Spain & Heriot-Watt University, Scotland
Thesis: "Real-time 2D Detection and Tracking of Construction Workers"
2008 - 2011
BSc. in Software Engineering and Management
Göteborgs Universitet, Sweden
Thesis: "Speech Recognition for Noisy Environments - Feasibility of Voice Command in Construction Settings"
Summer Schools
2016
Brains, Minds and Machines Summer Course, The Center for Brains, Minds and Machines (CBMM), Woods Hole, Massachusetts, United States
2017
Visual Neuroscience European Summer School, Rauischholzhausen, Germany
Ongoing Works
Arash Akbarinia, Karl R. Gegenfurtner, Paradox in Deep Neural Networks: Similar yet Different while Different yet Similar, arXiv preprint arXiv:1903.04772, 2019. (PDF)
Arash Akbarinia, Karl R. Gegenfurtner, Manifestation of Image Contrast in Deep Networks,
arXiv preprint arXiv:1902.04378, 2019. (PDF)
Arash Akbarinia, Karl R. Gegenfurtner, How is Contrast Encoded in Deep Neural Networks?, arXiv preprint arXiv:1809.01438, 2018. (PDF)
Publications
Journals
Arash Akbarinia, Karl R. Gegenfurtner, Color metamerism and the structure of illuminant space, Journal of the Optical Society of America A (JOSA A), 2018. (PDF)
Arash Akbarinia, C. Alejandro Párraga, Colour Constancy Beyond the Classical Receptive Field, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2017. (PDF) (Project Page)
Arash Akbarinia, C. Alejandro Párraga, Feedback and Surround Modulated Boundary Detection, International Journal of Computer Vision (IJCV), 2017. (PDF) (Project Page)
C. Alejandro Párraga, Arash Akbarinia, NICE: A Computational Solution to Close the Gap from Colour Perception to Colour Categorization, PLoS ONE, vol 11(3):1-32, 2016. (PDF) (Project Page)
Peer-reviewed Conferences
Xavier Soria, Angel D Sappa, Arash Akbarinia, Multispectral Single-Sensor RGB-NIR Imaging: New Challenges and Opportunities, in International Conference on Image Processing Theory, Tools and Applications (IPTA), Montreal, Canada, 2017. (PDF)
Arash Akbarinia, Raquel Gil Rodríguez, C. Alejandro Párraga, Colour Constancy: Biologically-inspired Contrast Variant Pooling Mechanism, in British Machine Vision Conference (BMVC), London, United Kingdom, 2017. (PDF)
Arash Akbarinia, C. Alejandro Párraga, Biologically Plausible Boundary Detection, in British Machine Vision Conference (BMVC), York, United Kingdom, 2016. (PDF) (Project Page)
Eduardo Tusa, Arash Akbarinia, Raquel Gil Rodríguez, Corina Barbalata, Real-Time Face Detection and Tracking Utilising OpenMP and ROS, in Asia-Pacific Conference on Computer Aided System Engineering (APCASE), Quito, Ecuador, 2015. (PDF)
Conferences
Arash Akbarinia, Karl R. Gegenfurtner, Contrast invariance in Deep Neural Networks (DNN), in European Conference on Visual Perception (ECVP), Trieste, Italy, 2018. (PDF)
Arash Akbarinia, Alejandro Párraga, Marta Expósito, Bogdan Raducanu, Xavier Otazu, Can biological solutions help computers detect symmetry?, in European Conference on Visual Perception (ECVP), Berlin, Germany, 2017. (PDF)
Arash Akbarinia, Karl R. Gegenfurtner, Metameric Mismatching in Natural and Artificial Reflectances, in Vision Sciences Society (VSS), Florida, United States of America, 2017. (PDF)
Arash Akbarinia, C. Alejandro Párraga, Dynamically Adjusted Surround Contrast Enhances Boundary Detection, in European Conference on Visual Perception (ECVP), Barcelona, Spain, 2016. (PDF)
C. Alejandro Párraga, Arash Akbarinia, Colour Constancy as a Product of Dynamic Centre-Surround Adaptation, in Vision Sciences Society (VSS), Florida, United States of America, 2016. (PDF)
Arash Akbarinia, C. Alejandro Párraga, Biologically Plausible Colour Naming Model, in European Conference on Visual Perception (ECVP), Liverpool, United Kingdom, 2015. (PDF) (Project Page)
Theses
Arash Akbarinia, Computational Model of Visual Perception: From Colour to Form, Universitat Autònoma de Barcelona, July 2017. (PDF) (Project Page)
Arash Akbarinia, Real-time 2D Detection and Tracking of Construction Workers, Erasmus Mundus in Vision and Robotics (VIBOT), Master Thesis, July 2013. (PDF) (Project Page)
Arash Akbarinia, Javier Valdez Medrano, Rashid Zamani, Speech Recognition for Noisy Environments-Feasibility of Voice Command in Construction Settings, University of Gothenburg, Bachelor Thesis, July 2011. (PDF) (Project Page)
