Posts

An original application of Optimal Transport theory for estimating stereo depth within a Transformer deep learning architecture.

Recently I stumbled upon a task where I had to find a mapping between a camera and a target plane, but with a little twist. I could define line equations in the plane, and once I had captured the image and found the corresponding lines, the problem boiled down to calculating the homography based on line correspondences.

Low-Rank Adaptation technique was introduced to adapt large language models to a specific downstream task while keeping the original model intact and, by doing that, dramatically reducing the number of trainbale parameters.