CLIP: Connecting Text and Images

This video explains how CLIP from OpenAI transforms Image Classification into a Text-Image similarity matching task. This is done with Contrastive Training and Zero-Shot Pattern-Exploiting Training. Thanks for watching!

Paper Links:
Clip (Blog Post): https://openai.com/blog/clip/
VirTex: https://arxiv.org/pdf/2006.06666.pdf
ConVIRT: https://arxiv.org/pdf/2010.00747.pdf
Pattern-Exploiting Training: https://arxiv.org/pdf/2001.07676.pdf
Vision Transformer (Blog Post, Nice Animation): https://ai.googleblog.com/2020/12/transformers-for-image-recognition-at.html

Thanks for watching! Please Subscribe!

Видео CLIP: Connecting Text and Images канала Henry AI Labs