Early-Stage Researcher @ L2S, CentraleSupelec, CNRS, University of Paris-Saclay - 12/2018-02/2024 - Saclay, France

Context

My thesis is part of a European Marie Skłodowska-Curie RealVision project. The aim is to train a new generation of scientists in the field of innovative hyperrealistic imaging technologies. Within the Signals and Systems Laboratory, a joint research unit (UMR 8506) of the CNRS, CentraleSupélec and Université Paris-Saclay, I lead research on learning-based coding for light field imaging.

Missions

  • Advance the coding frontier for light-field imaging.
  • Publish articles and scientific journals.
  • Collaborate with other RealVision project participants.

Results

  • The thesis defended in 2022 with one book chapter, three conference and one journal article published.
  • Proposed a hybrid coding scheme that combines a learning-based compression approach with a traditional video coding scheme for lossy compression of light field images.
  • Proposed a learning-based method for lossless light field compression.
  • Investigated the distributed source coding paradigm for light field compression.

Publications

  • Stepanov, M., 2022. Selected topics in learning‐based coding for light field imaging (Doctoral dissertation, Université Paris‐Saclay).
  • Stepanov, M., Valenzise, G. and Dufaux, F., 2022. Compression of light fields. In Immersive Video Technologies (pp. 201‐226).
  • Stepanov, M., Valenzise, G. and Dufaux, F., 2020, October. Hybrid Learning‐based and HEVC‐based Coding of Light Fields. In 2020 IEEE International Conference on Image Processing (ICIP) (pp. 3344‐3348). IEEE.
  • Mukati, M.U., Stepanov, M., Valenzise, G., Dufaux, F. and Forchhammer, S., 2020, July. View synthesis‐based distributed light field compression. In 2020 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) (pp. 1‐6). IEEE.
  • Mukati, M.U., Stepanov, M., Valenzise, G., Forchhammer, S. and Dufaux, F., 2021. Improved Deep Distributed Light Field Coding. IEEE Open Journal of Circuits and Systems, 2, pp.325‐337.
  • Stepanov, M., Mukati, M.U., Valenzise, G., Forchhammer, S. and Dufaux, F., 2021, October. Learning‐based lossless light field compression. In 2021 IEEE 23rd International Workshop on Multimedia Signal Processing (MMSP) (pp. 1‐6). IEEE

Technical Environment

Python, TensorFlow, PyTorch, OpenCV, sklearn, numpy, matplotlib, seaborn, TensorBoard, LaTeX, Git, MATLAB, Linux