After that, I turned out to Computer Vision and Pattern Recognition. I joined the LEAR group (INRIA Grenoble) as a permanent researcher in 2006, and moved to INRIA Rennes in 2009. I am not anymore at Inria.

My work is mainly focused on computer vision, with most of my contributions focused on large scale indexing techniques (vectors/images/video), focusing on compressed-domain search techniques that scale from millions to billions of vectors/images such as product quantization, that I proposed to search in a compressed set of vectors without decompressing it.