app.py

from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity
import gradio as gr

def predict(im1, im2):
  model = SentenceTransformer("sentence-transformers/clip-ViT-B-16")
  embeddings = model.encode([im1,im2])
  similarities = cosine_similarity(embeddings)
  sim = similarities[0][1]

  if sim > .8:
    return sim, "SAME PERSON, UNLOCK PHONE"
  else:
    return sim, "DIFFERENT PEOPLE, DON'T UNLOCK"
    
interface = gr.Interface(fn=predict, 
  inputs= [gr.Image(type="pil", source="webcam"), 
  gr.Image(type="pil", source="webcam")], 
  outputs= [gr.Number(label="Similarity"),
  gr.Textbox(label="Message")])
interface.launch(debug=True)