THUDM / CogVLM2

GPT4V-level open-source multi-modal model based on Llama3-8B
Apache License 2.0
1.42k stars 77 forks source link

CogVLM如何才能做图像中对象的定位任务呢? #82

Closed tiandazhao closed 3 weeks ago

tiandazhao commented 3 weeks ago

System Info / 系統信息

大家好,cogvlm2可以做对象标注任务吗?尝试过很多类型的prompt,都不能顺利输出图片的坐标,想问下,官方做在训练该类型任务时,有没有推荐的prompt啊

Who can help? / 谁可以帮助到您?

No response

Information / 问题信息

Reproduction / 复现过程

image

Expected behavior / 期待表现

期待能标出图片的中制定对象的坐标

zRzRzRzRzRzRzR commented 3 weeks ago

see #84