工 作 随 记

自监督篇

DINOv2: Learning Robust Visual Features without Supervision

https://github.com/facebookresearch/dinov2

多模态篇

OVSeg (+ Segment_Anything): Open-Vocabulary Semantic Segmentation with Mask-adapted CLIP

https://github.com/facebookresearch/ov-seg

Grounded-SAM: Marrying Grounding DINO with Segment Anything & Stable Diffusion & BLIP & Whisper - Automatically Detect , Segment and Generate Anything with Image, Text, and Speech Inputs.

https://github.com/IDEA-Research/Grounded-Segment-Anything

Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection

https://github.com/IDEA-Research/Grounded-Segment-Anything

BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation

https://github.com/salesforce/BLIP

LAVIS - A Library for Language-Vision Intelligence

https://github.com/salesforce/LAVIS

大模型篇

视觉篇

语言篇