recognize-anything
recognize-anything copied to clipboard
xinyu1205
Open-source and strong foundation image recognition models.
I collected the 4M datasets with some imgs urls failed (about 1.5%) , which causes 146 tags missing. I regenerate the label embedding, ram_tag_list_en(cn)/threshold according to the missing tags (get...
The current RAM can only tag images in a defined set of tags (or a customized set of tags), I wonder if the authors will consider generative tagging in the...
只输入一张图像，Tag2Text生成的caption并没有用上它生成全部的tags？此外，当Tag2Text的输入是一张图像和几个specific tags的时候，它生成的caption可能也并不包含specific tags？
Hello, I would like to finetune RAM++ tagging with other datasets. I spent a lot of time trying to understand how it works. But there are still quite a few...
Hello I want to fine tune the recognize-anything model to label images with tags for real people or cartoon characters. I have two questions: 1. Would fine tune just the...
Hi, @xinyu1205 After installing the package in different venv, I ran inference_ram_plus.py file and faced this error: `Traceback (most recent call last): File "/Users/pranavkumar/Desktop/py-image-tag/main.py", line 43, in model = ram_plus(pretrained=args.pretrained,...
Do you have an inference code for text in RAM? Tag2Text has text output(Image caption), but RAM only outputs tags. Can you help me out?
Is there any download link for 4M version checkpoint of Tag2Text, RAM and RAM++?
Hi, can we concat the label_embed between the default ram tag and open set / custom tag? thanks
Hi, In the dataset module, there is pre_caption function in utils.py. Why we need pre_caption? To align the dataset to have same length? Thanks