My Vision, Computer Vision

Learning to Prompt for Vision-Language ModelsLarge pre-trained vision-language models like CLIP have shown great potential in learning representations that are transferable across a wide range of downstream tasks. Different from the traditional representation learning that is based mostly on discretiarxiv.org발행일 : 2021년 9월 2일저널/학회 : SPRINGER 2022ProblemCLIP과 같은 기존 VLM은 프롬프트 엔지니어링을 통한 Zero-shot t..

CLIP-Adapter: Better Vision-Language Models with Feature AdaptersLarge-scale contrastive vision-language pre-training has shown significant progress in visual representation learning. Unlike traditional visual systems trained by a fixed set of discrete labels, a new paradigm was introduced in \cite{radford2021learning}arxiv.orgProblem(문제 지적)CLIP은 클래스 이름을 프롬프트 템플릿에 넣어 Zero-shot transfer를 수행하는데, 이..