|
import base64 |
|
from io import BytesIO |
|
|
|
import gradio as gr |
|
import torch |
|
from transformers import BlipForConditionalGeneration, BlipProcessor |
|
|
|
import modules.chat as chat |
|
import modules.shared as shared |
|
|
|
|
|
|
|
input_hijack = { |
|
'state': False, |
|
'value': ["", ""] |
|
} |
|
|
|
processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base") |
|
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base", torch_dtype=torch.float32).to("cpu") |
|
|
|
def caption_image(raw_image): |
|
inputs = processor(raw_image.convert('RGB'), return_tensors="pt").to("cpu", torch.float32) |
|
out = model.generate(**inputs, max_new_tokens=100) |
|
return processor.decode(out[0], skip_special_tokens=True) |
|
|
|
def generate_chat_picture(picture, name1, name2): |
|
text = f'*{name1} sends {name2} a picture that contains the following: "{caption_image(picture)}"*' |
|
buffer = BytesIO() |
|
picture.save(buffer, format="JPEG") |
|
img_str = base64.b64encode(buffer.getvalue()).decode('utf-8') |
|
visible_text = f'<img src="data:image/jpeg;base64,{img_str}">' |
|
return text, visible_text |
|
|
|
def ui(): |
|
picture_select = gr.Image(label='Send a picture', type='pil') |
|
|
|
function_call = 'chat.cai_chatbot_wrapper' if shared.args.cai_chat else 'chat.chatbot_wrapper' |
|
|
|
|
|
picture_select.upload(lambda picture, name1, name2: input_hijack.update({"state": True, "value": generate_chat_picture(picture, name1, name2)}), [picture_select, shared.gradio['name1'], shared.gradio['name2']], None) |
|
|
|
|
|
picture_select.upload(eval(function_call), shared.input_params, shared.gradio['display'], show_progress=shared.args.no_stream) |
|
|
|
|
|
picture_select.upload(lambda : None, [], [picture_select], show_progress=False) |
|
|