Research - Dmitry Kuzovkin

Research

My research path started with an interest in computer science and a passion for photography. Being curious about the underlying principles of digital image formation and processing, I pursued a Master's in this domain. In 2014, I received my master's degree in the CIMET programme, where I studied image processing, computer vision and color science. During this program I had an opportunity to study in Jean Monnet University (France), University of Granada (Spain) and Norwegian University of Science and Technology (Norway). In my final master thesis project, I worked on the topic of automatic example-based image colorization using descriptor matching.

In June 2019, I received a PhD degree. The doctoral studies were performed in agreement between University of Rennes, Technicolor company and IRISA lab. During my PhD, I worked with Tania Pouli, Rémi Cozot, Olivier Le Meur, Jonathan Kervec and Kadi Bouatouch. My thesis was focused on the subject of automatic photo assessment. I explored possible ways of photo evaluation based on image quality and aesthetics, and then I studied the influence of photo context in photo assessment, specifically the photo album context. I have conducted several user studies to investigate how different users group and select photos in photo collections. To model the nature of user decisions, I proposed a hierarchical clustering method to cluster similar photos in a multi-level similarity structure and a context-adaptive photo assessment model using a machine learning approach. In addition, during my PhD I became largely interested in deep learning, particularly in GAN networks and their potential creative applications in photography.

Publications

  • Context in photo albums: understanding and modeling user behavior in clustering and selection [pdf]
    Kuzovkin, D., Pouli, T., Le Meur, O., Cozot, R., Kervec, J. and Bouatouch, K., 2019, August.
    In ACM TAP 2019: ACM Transactions on Applied Perception (TAP), 16(2), p.11.
  • Image Selection in Photo Albums [pdf]
    Kuzovkin, D., Pouli, T., Cozot, R., Le Meur, O., Kervec, J. and Bouatouch, K., 2018, June.
    In ACM ICMR 2018: Proceedings of the 2018 ACM on International Conference on Multimedia Retrieval (pp. 397-404).
  • Context-aware clustering and assessment of photo collections [pdf]
    Kuzovkin, D., Pouli, T., Cozot, R., Le Meur, O., Kervec, J. and Bouatouch, K., 2017, July.
    In ACM Expressive 2017: Proceedings of the symposium on Computational Aesthetics (p. 6).
  • Colour mapping: A review of recent methods, extensions and applications [pdf]
    Faridul, H.S., Pouli, T., Chamaret, C., Stauder, J., Reinhard, E., Kuzovkin, D. and Trémeau, A., 2016, February.
    In CGF 2016: Computer Graphics Forum (Vol. 35, No. 1, pp. 59-88).
  • Descriptor-based image colorization and regularization [pdf]
    Kuzovkin, D., Chamaret, C. and Pouli, T., 2015, March.
    In CCIW 2015: International Workshop on Computational Color Imaging (pp. 59-68). Springer, Cham.

PhD Thesis

Assessment of photos in albums based on aesthetics and context [pdf]
Dmitry Kuzovkin, 2019.