Computer Vision from Small Beginnings to VLMs

​Description:
This talk will trace computer vision's journey from early edge detection and handcrafted features to today's powerful Vision Language Models. I'll cover key milestones including CNNs, ImageNet's impact, Vision Transformers, the rise of self-supervised learning in computer vision, CLIP, image generation and finally the vision language models and their current state

​Speaker:
Irakli Salia is working on building advanced computer vision applications at EPAM Systems ranging from classical computer vision problems (such as object detection, segmentation) to generative ones. He is also an active contributor to PyTorch ecosystem ranging from writing specialized kernels on MPS as well as fixing various issues in the library.

​Location:
Rooms Hotel Tbilisi, Central Room -1 floor (The main entrance, Chovelidze street side)