search

kennethwdk / LocLLM

Code for "LocLLM: Exploiting Generalizable Human Keypoint Localization via Large Language Model", CVPR 2024 Highlight
MIT License
31 stars 3 forks source link

first stage pre-training #4

Open starhiking opened 4 months ago

starhiking commented 4 months ago

Hi author,

It is wonderful work, Thank you for your code and models.

Can you release the first stage pre-training shell?

starhiking commented 4 months ago

Also, can you provide the ckpt of the first-stage linear?

KangqingYe commented 2 weeks ago

I found the pretrained projection layer provided by LLaVA at the following links: https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md and https://huggingface.co/liuhaotian/llava-v1.5-mlp2x-336px-pretrain-vicuna-7b-v1.5/tree/main, but I am uncertain whether this is the version used by the author.