Update app.py

app.py

@@ -1,183 +1,253 @@
+import os
+import cv2
+from PIL import Image
 import gradio as gr
 import numpy as np
 import random
-from diffusers import DiffusionPipeline
-import torch
-
-device = "cuda" if torch.cuda.is_available() else "cpu"
-
-if torch.cuda.is_available():
-    torch.cuda.max_memory_allocated(device=device)
-    pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.float16, variant="fp16", use_safetensors=True)
-    pipe.enable_xformers_memory_efficient_attention()
-    pipe = pipe.to(device)
-else: 
-    pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", use_safetensors=True)
-    pipe = pipe.to(device)
-
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
-
-def infer(prompt_part1, color, dress_type, design, prompt_part5, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps):
-    prompt = f"{prompt_part1} {color} colored plain {dress_type} with {design} design, {prompt_part5}"
-    
+import base64
+import requests
+import json
+import time
+from requests.adapters import HTTPAdapter
+
+def tryon(person_img, garment_img, seed, randomize_seed):
+    post_start_time = time.time()
+    if person_img is None or garment_img is None:
+        return None, None, "Empty image"
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
-        
-    generator = torch.Generator().manual_seed(seed)
-    
-    image = pipe(
-        prompt=prompt, 
-        negative_prompt=negative_prompt,
-        guidance_scale=guidance_scale, 
-        num_inference_steps=num_inference_steps, 
-        width=width, 
-        height=height,
-        generator=generator
-    ).images[0] 
-    
-    return image
-
-examples = [
-    "red, t-shirt, yellow stripes",
-    "blue, hoodie, minimalist",
-    "red, sweat shirt, geometric design",
-]
-
-css = """
-#col-container {
+    encoded_person_img = cv2.imencode('.jpg', cv2.cvtColor(person_img, cv2.COLOR_RGB2BGR))[1].tobytes()
+    encoded_person_img = base64.b64encode(encoded_person_img).decode('utf-8')
+    encoded_garment_img = cv2.imencode('.jpg', cv2.cvtColor(garment_img, cv2.COLOR_RGB2BGR))[1].tobytes()
+    encoded_garment_img = base64.b64encode(encoded_garment_img).decode('utf-8')
+
+    url = "http://" + os.environ['tryon_url'] + "Submit"
+    token = os.environ['token']
+    cookie = os.environ['Cookie']
+    referer = os.environ['referer']
+    headers = {'Content-Type': 'application/json', 'token': token, 'Cookie': cookie, 'referer': referer}
+    data = {
+        "clothImage": encoded_garment_img,
+        "humanImage": encoded_person_img,
+        "seed": seed
+    }
+    try:
+        response = requests.post(url, headers=headers, data=json.dumps(data), timeout=50)
+        print("post response code", response.status_code)
+        if response.status_code == 200:
+            result = response.json()['result']
+            status = result['status']
+            if status == "success":
+                uuid = result['result']
+                print(uuid)
+    except Exception as err:
+        print(f"Error: {err}")
+        raise gr.Error("Too many users, please try again later")
+    post_end_time = time.time()
+    print(f"post time used: {post_end_time-post_start_time}")
+
+    get_start_time =time.time()
+    time.sleep(9)
+    Max_Retry = 10
+    result_img = None
+    for i in range(Max_Retry):
+        try:
+            url = "http://" + os.environ['tryon_url'] + "Query?taskId=" + uuid
+            response = requests.get(url, headers=headers, timeout=15)
+            print("get response code", response.status_code)
+            if response.status_code == 200:
+                result = response.json()['result']
+                status = result['status']
+                if status == "success":
+                    result = base64.b64decode(result['result'])
+                    result_np = np.frombuffer(result, np.uint8)
+                    result_img = cv2.imdecode(result_np, cv2.IMREAD_UNCHANGED)
+                    result_img = cv2.cvtColor(result_img, cv2.COLOR_RGB2BGR)
+                    info = "Success"
+                    break
+                elif status == "error":
+                    raise gr.Error("Too many users, please try again later")
+            else:
+                print(response.text)
+                info = "URL error, pleace contact the admin"
+        except requests.exceptions.ReadTimeout:
+            print("timeout")
+            info = "Too many users, please try again later"
+        time.sleep(1)
+    get_end_time = time.time()
+    print(f"get time used: {get_end_time-get_start_time}")
+
+    return result_img, seed, info
+
+def start_tryon(person_img, garment_img, seed, randomize_seed):
+    start_time = time.time()
+    if person_img is None or garment_img is None:
+        return None, None, "Empty image"
+    if randomize_seed:
+        seed = random.randint(0, MAX_SEED)
+    encoded_person_img = cv2.imencode('.jpg', cv2.cvtColor(person_img, cv2.COLOR_RGB2BGR))[1].tobytes()
+    encoded_person_img = base64.b64encode(encoded_person_img).decode('utf-8')
+    encoded_garment_img = cv2.imencode('.jpg', cv2.cvtColor(garment_img, cv2.COLOR_RGB2BGR))[1].tobytes()
+    encoded_garment_img = base64.b64encode(encoded_garment_img).decode('utf-8')
+
+    url = "http://" + os.environ['tryon_url']
+    token = os.environ['token']
+    cookie = os.environ['Cookie']
+    referer = os.environ['referer']
+
+    headers = {'Content-Type': 'application/json', 'token': token, 'Cookie': cookie, 'referer': referer}
+    data = {
+        "clothImage": encoded_garment_img,
+        "humanImage": encoded_person_img,
+        "seed": seed
+    }
+
+    result_img = None
+    try:
+        session = requests.Session()
+        response = session.post(url, headers=headers, data=json.dumps(data), timeout=60)
+        print("response code", response.status_code)
+        if response.status_code == 200:
+            result = response.json()['result']
+            status = result['status']
+            if status == "success":
+                result = base64.b64decode(result['result'])
+                result_np = np.frombuffer(result, np.uint8)
+                result_img = cv2.imdecode(result_np, cv2.IMREAD_UNCHANGED)
+                result_img = cv2.cvtColor(result_img, cv2.COLOR_RGB2BGR)
+                info = "Success"
+            else:
+                info = "Try again latter"
+        else:
+            print(response.text)
+            info = "URL error, pleace contact the admin"
+    except requests.exceptions.ReadTimeout:
+        print("timeout")
+        info = "Too many users, please try again later"
+        raise gr.Error("Too many users, please try again later")
+    except Exception as err:
+        print(f"其他错误: {err}")
+        info = "Error, pleace contact the admin"
+    end_time = time.time()
+    print(f"time used: {end_time-start_time}")
+
+    return result_img, seed, info
+
+MAX_SEED = 999999
+
+example_path = os.path.join(os.path.dirname(__file__), 'assets')
+
+garm_list = os.listdir(os.path.join(example_path,"cloth"))
+garm_list_path = [os.path.join(example_path,"cloth",garm) for garm in garm_list]
+
+human_list = os.listdir(os.path.join(example_path,"human"))
+human_list_path = [os.path.join(example_path,"human",human) for human in human_list]
+
+css="""
+#col-left {
     margin: 0 auto;
-    max-width: 520px;
+    max-width: 430px;
+}
+#col-mid {
+    margin: 0 auto;
+    max-width: 430px;
+}
+#col-right {
+    margin: 0 auto;
+    max-width: 430px;
+}
+#col-showcase {
+    margin: 0 auto;
+    max-width: 1100px;
+}
+#button {
+    color: blue;
 }
 """
 
-if torch.cuda.is_available():
-    power_device = "GPU"
-else:
-    power_device = "CPU"
-
-with gr.Blocks(css=css) as demo:
-    
-    with gr.Column(elem_id="col-container"):
-        gr.Markdown(f"""
-        # Text-to-Image Gradio Template
-        Currently running on {power_device}.
-        """)
-        
-        with gr.Row():
-            
-            prompt_part1 = gr.Textbox(
-                value="a single", 
-                label="Prompt Part 1",
-                show_label=False,
-                interactive=False,
-                container=False,
-                elem_id="prompt_part1",
-                visible=False,
-            )
-            
-            prompt_part2 = gr.Textbox(
-                label="color",
-                show_label=False,
-                max_lines=1,
-                placeholder="color (e.g., color category)",
-                container=False,
-            )
-            
-            prompt_part3 = gr.Textbox(
-                label="dress_type",
-                show_label=False,
-                max_lines=1,
-                placeholder="dress_type (e.g., t-shirt, sweatshirt, shirt, hoodie)",
-                container=False,
-            )
-            
-            prompt_part4 = gr.Textbox(
-                label="design",
-                show_label=False,
-                max_lines=1,
-                placeholder="design",
-                container=False,
-            )
-            
-            prompt_part5 = gr.Textbox(
-                value="hanging on the plain wall", 
-                label="Prompt Part 5",
-                show_label=False,
-                interactive=False,
-                container=False,
-                elem_id="prompt_part5",
-                visible=False,
-            )
-            
-            run_button = gr.Button("Run", scale=0)
-        
-        result = gr.Image(label="Result", show_label=False)
-
-        with gr.Accordion("Advanced Settings", open=False):
-            
-            negative_prompt = gr.Textbox(
-                label="Negative prompt",
-                max_lines=1,
-                placeholder="Enter a negative prompt",
-                visible=False,
+def load_description(fp):
+    with open(fp, 'r', encoding='utf-8') as f:
+        content = f.read()
+    return content
+
+def change_imgs(image1, image2):
+    return image1, image2
+
+with gr.Blocks(css=css) as Tryon:
+    gr.HTML(load_description("assets/title.md"))
+    with gr.Row():
+        with gr.Column(elem_id = "col-left"):
+            gr.HTML("""
+            <div style="display: flex; justify-content: center; align-items: center; text-align: center; font-size: 20px;">
+                <div>
+                Step 1.  Upload a person image ⬇️
+                </div>
+            </div>
+            """)
+        with gr.Column(elem_id = "col-mid"):
+            gr.HTML("""
+            <div style="display: flex; justify-content: center; align-items: center; text-align: center; font-size: 20px;">
+                <div>
+                Step 2. Upload a garment image ⬇️
+                </div>
+            </div>
+            """)
+        with gr.Column(elem_id = "col-right"):
+            gr.HTML("""
+            <div style="display: flex; justify-content: center; align-items: center; text-align: center; font-size: 20px;">
+                <div>
+                Step 3. Press “Run” to get try-on results
+                </div>
+            </div>
+            """)
+    with gr.Row():
+        with gr.Column(elem_id = "col-left"):
+            imgs = gr.Image(label="Person image", sources='upload', type="numpy")
+            # category = gr.Dropdown(label="Garment category", choices=['upper_body', 'lower_body', 'dresses'],  value="upper_body")
+            example = gr.Examples(
+                inputs=imgs,
+                examples_per_page=12,
+                examples=human_list_path
             )
-            
-            seed = gr.Slider(
-                label="Seed",
-                minimum=0,
-                maximum=MAX_SEED,
-                step=1,
-                value=0,
+        with gr.Column(elem_id = "col-mid"):
+            garm_img = gr.Image(label="Garment image", sources='upload', type="numpy")
+            example = gr.Examples(
+                inputs=garm_img,
+                examples_per_page=12,
+                examples=garm_list_path
             )
-            
-            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
-            
+        with gr.Column(elem_id = "col-right"):
+            image_out = gr.Image(label="Result", show_share_button=False)
             with gr.Row():
-                
-                width = gr.Slider(
-                    label="Width",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=512,
-                )
-                
-                height = gr.Slider(
-                    label="Height",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=512,
-                )
-            
-            with gr.Row():
-                
-                guidance_scale = gr.Slider(
-                    label="Guidance scale",
-                    minimum=0.0,
-                    maximum=10.0,
-                    step=0.1,
-                    value=0.0,
-                )
-                
-                num_inference_steps = gr.Slider(
-                    label="Number of inference steps",
-                    minimum=1,
-                    maximum=12,
+                seed = gr.Slider(
+                    label="Seed",
+                    minimum=0,
+                    maximum=MAX_SEED,
                     step=1,
-                    value=2,
+                    value=0,
                 )
-        
-        gr.Examples(
-            examples=examples,
-            inputs=[prompt_part2]
-        )
-
-    run_button.click(
-        fn=infer,
-        inputs=[prompt_part1, prompt_part2, prompt_part3, prompt_part4, prompt_part5, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
-        outputs=[result]
-    )
-
-demo.queue().launch()
+                randomize_seed = gr.Checkbox(label="Random seed", value=True)
+            with gr.Row():
+                seed_used = gr.Number(label="Seed used")
+                result_info = gr.Text(label="Response")
+            # try_button = gr.Button(value="Run", elem_id="button")
+            test_button = gr.Button(value="Run", elem_id="button")
+
+
+    # try_button.click(fn=start_tryon, inputs=[imgs, garm_img, seed, randomize_seed], outputs=[image_out, seed_used, result_info], api_name='tryon',concurrency_limit=10)
+    test_button.click(fn=tryon, inputs=[imgs, garm_img, seed, randomize_seed], outputs=[image_out, seed_used, result_info], api_name='tryon',concurrency_limit=10)
+
+    with gr.Column(elem_id = "col-showcase"):
+        gr.HTML("""
+        <div style="display: flex; justify-content: center; align-items: center; text-align: center; font-size: 20px;">
+            <div> </div>
+            <br>
+            <div>
+            Virtual try-on examples in pairs of person and garment images
+            </div>
+        </div>
+        """)
+
+# ip = requests.get('http://ifconfig.me/ip', timeout=1).text.strip()
+# print("ip address", ip)
+Tryon.queue(max_size = 20).launch(max_threads = 5)