The AFRIF 2021 thesis prize awarded to Ignacio Rocco

The AFRIF (French Association for the Recognition and Interpretation of Forms) awards its thesis prize each year. The objective of this award is to highlight and encourage the best doctoral work on these subjects and to boost research in the field. The AFRIF 2021 thesis prize was awarded to Ignacio Rocco and a special mention was made to Hugo Richard.

AFRIF recently presented the winner of its 2021 thesis prize. The finalists had to have defended their thesis in the fields of image recognition between July 1, 2020 and December 31, 2021 in a French doctoral school or within the framework of joint supervision with a French doctoral school. The theses were evaluated by a jury chaired by Vincent Lepetit (ENPC, Paris) and made up of members of the board of directors, including Marie-Odile Berger (INRIA Nancy – Grand Est), president of AFRIF.

AFRIF 2021 thesis prize awarded to Ignacio Rocco

The AFRIF 2021 thesis prize was awarded to Ignacio Rocco for his work entitled “ Neural Architectures for Estimating Correspondences Between Images », carried out at the Ecole Normale Supérieure – Université PSL and under the direction of Josef Sivic and Relja Arandjelović. The thesis deals with the development of methods for matching between pairs of images in difficult situations such as extreme change of lighting, scenes with little texture or including repetitive structures or matching between parts of images. ‘objects that belong to the same class but may have large intra-class differences in appearance. Ignacio Rocco contributes as follows:

  1. Develop a trainable approach for parametric image alignment using a Siamese network model
  2. Design a weakly supervised training approach that allows training from real image pairs annotated only at the image pair level
  3. Propose Neighborhood Consensus Networks that can be used for the purpose of robustly estimating matches for tasks where discrete matches are required
  4. Develop a more efficient variant capable of reducing the memory requirements and execution time of Neighborhood Consensus Networks by a factor of ten

Special mention for Hugo Richard

A special mention was also made to Hugo Richard for his work entitled ” Unsupervised Component Analysis for Neuroimaging Data », carried out at the University of Paris-Saclay under the direction of Bertrand Thirion. This work is a computer science and mathematics thesis that applies to the field of neuroscience and more particularly to research on the modeling of human cerebral activity by electrophysiology and imaging. The need to use mathematical tools stems from the complexity of identifying neuronal activity from data collected from so-called natural stimuli. In this thesis, the author first considers the case of shared response template, in which subjects are assumed to share a common response. It is useful for reducing the dimension of the data but its training remains nevertheless expensive for the functional imaging data whose dimension can be immense. This model also makes unrealistic assumptions about the above data. Another method giving more realistic assumptions is independent component analysis, but it is difficult to generalize to datasets that contain multiple topics. Hugo Richard therefore proposes an extension of the latter which he calls Multi-view ACIit is based on the maximum likelihood principle which is suitable for multi-subject datasets.

The first AFRIF prize for the 2021 edition was awarded €1,000 and will be invited to present their work as part of the RFIAP 2022 young researchers conference to be held in Vannes from July 5 to 8, 2022.

We wish to say thanks to the writer of this article for this awesome content

The AFRIF 2021 thesis prize awarded to Ignacio Rocco

Explore our social media accounts as well as other pages related to them