app.py

import streamlit as st
import open_clip
import torch
import requests
from PIL import Image
from io import BytesIO
import time
import json
import numpy as np
from ultralytics import YOLO
import cv2

# Load CLIP model and tokenizer
@st.cache_resource
def load_clip_model():
    model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:Marqo/marqo-fashionSigLIP')
    tokenizer = open_clip.get_tokenizer('hf-hub:Marqo/marqo-fashionSigLIP')
    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
    model.to(device)
    return model, preprocess_val, tokenizer, device

clip_model, preprocess_val, tokenizer, device = load_clip_model()

# Load YOLOv8 model
@st.cache_resource
def load_yolo_model():
    return YOLO("./best.pt")

yolo_model = load_yolo_model()

# Load and process data
@st.cache_data
def load_data():
    with open('./musinsa-final.json', 'r', encoding='utf-8') as f:
        return json.load(f)

data = load_data()

# Helper functions
def load_image_from_url(url, max_retries=3):
    for attempt in range(max_retries):
        try:
            response = requests.get(url, timeout=10)
            response.raise_for_status()
            img = Image.open(BytesIO(response.content)).convert('RGB')
            return img
        except (requests.RequestException, Image.UnidentifiedImageError) as e:
            if attempt < max_retries - 1:
                time.sleep(1)
            else:
                return None

def get_image_embedding(image):
    image_tensor = preprocess_val(image).unsqueeze(0).to(device)
    with torch.no_grad():
        image_features = clip_model.encode_image(image_tensor)
        image_features /= image_features.norm(dim=-1, keepdim=True)
    return image_features.cpu().numpy()

@st.cache_data
def process_database():
    database_embeddings = []
    database_info = []

    for item in data:
        image_url = item['이미지 링크'][0]
        image = load_image_from_url(image_url)
        if image is not None:
            embedding = get_image_embedding(image)
            database_embeddings.append(embedding)
            database_info.append({
                'id': item['\ufeff상품 ID'],
                'category': item['카테고리'],
                'brand': item['브랜드명'],
                'name': item['제품명'],
                'price': item['정가'],
                'discount': item['할인율'],
                'image_url': image_url
            })
        else:
            st.warning(f"Skipping item {item['상품 ID']} due to image loading failure")

    if database_embeddings:
        return np.vstack(database_embeddings), database_info
    else:
        st.error("No valid embeddings were generated.")
        return None, None

database_embeddings, database_info = process_database()

def get_text_embedding(text):
    text_tokens = tokenizer([text]).to(device)
    with torch.no_grad():
        text_features = clip_model.encode_text(text_tokens)
        text_features /= text_features.norm(dim=-1, keepdim=True)
    return text_features.cpu().numpy()

def find_similar_images(query_embedding, top_k=5):
    similarities = np.dot(database_embeddings, query_embedding.T).squeeze()
    top_indices = np.argsort(similarities)[::-1][:top_k]
    results = []
    for idx in top_indices:
        results.append({
            'info': database_info[idx],
            'similarity': similarities[idx]
        })
    return results

def detect_clothing(image):
    results = yolo_model(image)
    detections = results[0].boxes.data.cpu().numpy()
    categories = []
    for detection in detections:
        x1, y1, x2, y2, conf, cls = detection
        category = yolo_model.names[int(cls)]
        if category in ['sunglass','hat','jacket','shirt','pants','shorts','skirt','dress','bag','shoe']:
            categories.append({
                'category': category,
                'bbox': [int(x1), int(y1), int(x2), int(y2)],
                'confidence': conf
            })
    return categories

def crop_image(image, bbox):
    return image.crop((bbox[0], bbox[1], bbox[2], bbox[3]))

# 세션 상태 초기화
if 'step' not in st.session_state:
    st.session_state.step = 'input'
if 'query_image_url' not in st.session_state:
    st.session_state.query_image_url = ''
if 'detections' not in st.session_state:
    st.session_state.detections = []
if 'selected_category' not in st.session_state:
    st.session_state.selected_category = None

# Streamlit app
st.title("Advanced Fashion Search App")

# 단계별 처리
if st.session_state.step == 'input':
    st.session_state.query_image_url = st.text_input("Enter image URL:", st.session_state.query_image_url)
    if st.button("Detect Clothing"):
        if st.session_state.query_image_url:
            query_image = load_image_from_url(st.session_state.query_image_url)
            if query_image is not None:
                st.session_state.query_image = query_image
                st.session_state.detections = detect_clothing(query_image)
                if st.session_state.detections:
                    st.session_state.step = 'select_category'
                else:
                    st.warning("No clothing items detected in the image.")
            else:
                st.error("Failed to load the image. Please try another URL.")
        else:
            st.warning("Please enter an image URL.")

# Update the 'select_category' step
elif st.session_state.step == 'select_category':
    st.image(st.session_state.query_image, caption="Query Image", use_column_width=True)
    st.subheader("Detected Clothing Items:")
    
    for detection in st.session_state.detections:
        col1, col2 = st.columns([1, 3])
        with col1:
            st.write(f"{detection['category']} (Confidence: {detection['confidence']:.2f})")
        with col2:
            cropped_image = crop_image(st.session_state.query_image, detection['bbox'])
            st.image(cropped_image, caption=detection['category'], use_column_width=True)
    
    options = [f"{d['category']} (Confidence: {d['confidence']:.2f})" for d in st.session_state.detections]
    selected_option = st.selectbox("Select a category to search:", options)
    
    if st.button("Search Similar Items"):
        st.session_state.selected_category = selected_option
        st.session_state.step = 'show_results'

elif st.session_state.step == 'show_results':
    st.image(st.session_state.query_image, caption="Query Image", use_column_width=True)
    selected_detection = next(d for d in st.session_state.detections 
                              if f"{d['category']} (Confidence: {d['confidence']:.2f})" == st.session_state.selected_category)
    cropped_image = crop_image(st.session_state.query_image, selected_detection['bbox'])
    st.image(cropped_image, caption="Cropped Image", use_column_width=True)
    query_embedding = get_image_embedding(cropped_image)
    similar_images = find_similar_images(query_embedding)
    
    st.subheader("Similar Items:")
    for img in similar_images:
        col1, col2 = st.columns(2)
        with col1:
            st.image(img['info']['image_url'], use_column_width=True)
        with col2:
            st.write(f"Name: {img['info']['name']}")
            st.write(f"Brand: {img['info']['brand']}")
            st.write(f"Category: {img['info']['category']}")
            st.write(f"Price: {img['info']['price']}")
            st.write(f"Discount: {img['info']['discount']}%")
            st.write(f"Similarity: {img['similarity']:.2f}")
    
    if st.button("Start New Search"):
        st.session_state.step = 'input'
        st.session_state.query_image_url = ''
        st.session_state.detections = []
        st.session_state.selected_category = None

else:  # Text search
    query_text = st.text_input("Enter search text:")
    if st.button("Search by Text"):
        if query_text:
            text_embedding = get_text_embedding(query_text)
            similar_images = find_similar_images(text_embedding)
            st.subheader("Similar Items:")
            for img in similar_images:
                col1, col2 = st.columns(2)
                with col1:
                    st.image(img['info']['image_url'], use_column_width=True)
                with col2:
                    st.write(f"Name: {img['info']['name']}")
                    st.write(f"Brand: {img['info']['brand']}")
                    st.write(f"Category: {img['info']['category']}")
                    st.write(f"Price: {img['info']['price']}")
                    st.write(f"Discount: {img['info']['discount']}%")
                    st.write(f"Similarity: {img['similarity']:.2f}")
        else:
            st.warning("Please enter a search text.")