AlphaCLIP icon indicating copy to clipboard operation
AlphaCLIP copied to clipboard

[CVPR 2024] Alpha-CLIP: A CLIP Model Focusing on Wherever You Want

Results 14 AlphaCLIP issues
Sort by recently updated
recently updated
newest added

Hi, congratulations on this awesome paper. Are you planning on finetuning the model on the CLIP VIT-H/14 architecture in the near future. Thanks in advance.

UnpicklingError Traceback (most recent call last) [](https://localhost:8080/#) in () 7 # load model and prepare mask transform 8 device = "cuda" if torch.cuda.is_available() else "cpu" ----> 9 model, preprocess =...

when I run the app.py in demo/with_llm, I find the "send" button is useless. I have download all the required checkpoints. This is my terminal error: ![image](https://github.com/SunzeY/AlphaCLIP/assets/129270219/e03bd209-55f6-40ac-b2b4-ee8c4a49bb6a) ![image](https://github.com/SunzeY/AlphaCLIP/assets/129270219/6238e34a-627a-4a6e-9c75-df15cc74dc4d)

Demo i deployed locally didn't work. There is no output in the terminal. ![image](https://github.com/SunzeY/AlphaCLIP/assets/76839625/e4c1c479-dd86-4eda-a507-32dbe1482aec)

Thank you for your work. Considering the captions in the GRIT dataset consist solely of noun words like _berries_, _person_ ... Did you use **Templates** to expand the captions, such...