Vision-Language Model Guided Training
第四阶段第六次会议 主讲人:马文轩
VLM guided training
大纲
Object Detection
- ViLD (two-stage, open vocabulary)
- DetPro (ViLD+prompt learning)
- HierKD (one-stage, open vocabulary)
- Pix2Seq (language modeling framework)
Image Segmentation
- Lseg (open vocabulary)
- CLIMS (weak-supervised segmentation)
Other applications
- GALS (attention)
- PLG (deep metric learning)
Talk 视频回放
https://meeting.tencent.com/v2/cloud-record/share?id=a44ef18e-b57f-4c0d-90a9-14f9ffe20415&from=3
Slide
链接:https://pan.baidu.com/s/1vJ9S2Cv-YLpN_w4aefq6tg 提取码:o1f6