File size: 9,841 Bytes
69620c8
 
 
 
 
 
6960db5
54d0511
31c0b50
 
 
 
6960db5
 
 
8c89a89
69620c8
8c89a89
 
69620c8
8c89a89
c1bd24e
 
69620c8
 
8c89a89
 
 
 
 
 
 
 
 
 
 
 
 
c401dbb
8c89a89
6960db5
 
 
 
 
 
 
 
 
 
8c89a89
 
 
 
 
 
 
 
69620c8
 
 
 
8c89a89
69620c8
 
4f076f3
69620c8
 
 
6960db5
69620c8
 
 
 
 
 
 
 
 
4f076f3
69620c8
ab309a4
c0fd987
26c3a03
8c89a89
 
f207d6b
c0fd987
 
 
 
 
 
 
f207d6b
c0fd987
8c89a89
c0fd987
 
8c89a89
 
f207d6b
c0fd987
 
8c89a89
c0fd987
8c89a89
c0fd987
 
f207d6b
c0fd987
8c89a89
 
 
 
 
 
 
 
f207d6b
26c3a03
c0fd987
26c3a03
f207d6b
c0fd987
f207d6b
c0fd987
 
 
 
f207d6b
c0fd987
26c3a03
 
c0fd987
26c3a03
 
c0fd987
 
8c89a89
c0fd987
8c89a89
c0fd987
 
 
 
8c89a89
 
c0fd987
 
8c89a89
 
c0fd987
 
 
8c89a89
 
 
c0fd987
 
f207d6b
c0fd987
 
 
f207d6b
c0fd987
8c89a89
 
 
c0fd987
8c89a89
63f9491
26c3a03
8c89a89
 
 
 
 
c0fd987
63f9491
c0fd987
4b8e8f5
 
 
 
 
 
 
 
 
 
 
 
 
2cb92de
4b8e8f5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2cb92de
4b8e8f5
 
 
 
 
 
 
 
 
 
 
 
 
2cb92de
4b8e8f5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c0fd987
 
4b8e8f5
 
 
 
 
 
69620c8
8c89a89
 
69620c8
 
 
4b8e8f5
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
import spaces
import gradio as gr
import torch
from PIL import Image
from diffusers import DiffusionPipeline
import random
from transformers import pipeline

torch.backends.cudnn.deterministic = True
torch.backends.cudnn.benchmark = False
torch.backends.cuda.matmul.allow_tf32 = True

# ๋ฒˆ์—ญ ๋ชจ๋ธ ์ดˆ๊ธฐํ™”
translator = pipeline("translation", model="Helsinki-NLP/opus-mt-ko-en")

# ๊ธฐ๋ณธ ๋ชจ๋ธ ๋ฐ LoRA ์„ค์ •
base_model = "black-forest-labs/FLUX.1-dev"
model_lora_repo = "Motas/Flux_Fashion_Photography_Style"  # ํŒจ์…˜ ๋ชจ๋ธ LoRA
clothes_lora_repo = "prithivMLmods/Canopus-Clothing-Flux-LoRA"  # ์˜๋ฅ˜ LoRA

pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
pipe.to("cuda")

MAX_SEED = 2**32-1

# ์˜ˆ์ œ ์„ค์ •
example_model_prompts = [
    "professional fashion model, full body shot, standing pose, natural lighting, studio background, high fashion, elegant pose",
    "fashion model portrait, upper body, confident pose, fashion photography, neutral background, professional lighting",
    "stylish fashion model, three-quarter view, editorial pose, high-end fashion magazine style, minimal background"
]

example_clothes_prompts = [
    "luxury designer sweater, cashmere material, cream color, cable knit pattern, high-end fashion, product photography",
    "elegant business blazer, tailored fit, charcoal grey, premium wool fabric, professional wear",
    "modern streetwear hoodie, oversized fit, minimalist design, premium cotton, urban style"
]

@spaces.GPU()
def generate_fashion(prompt, mode, cfg_scale, steps, randomize_seed, seed, width, height, lora_scale, progress=gr.Progress(track_tqdm=True)):
    # ํ•œ๊ธ€ ๊ฐ์ง€ ๋ฐ ๋ฒˆ์—ญ
    def contains_korean(text):
        return any(ord('๊ฐ€') <= ord(char) <= ord('ํžฃ') for char in text)
    
    if contains_korean(prompt):
        translated = translator(prompt)[0]['translation_text']
        actual_prompt = translated
    else:
        actual_prompt = prompt

    # ๋ชจ๋“œ์— ๋”ฐ๋ฅธ LoRA ๋ฐ ํŠธ๋ฆฌ๊ฑฐ์›Œ๋“œ ์„ค์ •
    if mode == "Generate Model":
        pipe.load_lora_weights(model_lora_repo)
        trigger_word = "fashion photography, professional model"
    else:
        pipe.load_lora_weights(clothes_lora_repo)
        trigger_word = "upper clothing, fashion item"

    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    generator = torch.Generator(device="cuda").manual_seed(seed)

    progress(0, "Starting fashion generation...")

    for i in range(1, steps + 1):
        if i % (steps // 10) == 0:
            progress(i / steps * 100, f"Processing step {i} of {steps}...")

    image = pipe(
        prompt=f"{actual_prompt} {trigger_word}",
        num_inference_steps=steps,
        guidance_scale=cfg_scale,
        width=width,
        height=height,
        generator=generator,
        joint_attention_kwargs={"scale": lora_scale},
    ).images[0]

    progress(100, "Completed!")
    return image, seed

# CSS ์ •์˜
custom_css = """
body {
    background-color: #f5f5f5;
    font-family: 'Arial', sans-serif;
}

.container {
    max-width: 1200px;
    margin: 0 auto;
    padding: 20px;
}

.header {
    text-align: center;
    color: #333;
    margin-bottom: 30px;
    font-size: 2.5em;
    text-transform: uppercase;
    letter-spacing: 2px;
}

.box-common {
    background-color: white;
    border-radius: 15px;
    padding: 25px;
    margin: 20px 0;
    box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
}

.mode-box {
    background-color: #fff;
    padding: 20px;
    border-radius: 10px;
    margin-bottom: 20px;
    border: 2px solid #eee;
}

.result-box {
    width: 90%;
    max-width: 1000px;
    margin: 20px auto;
}

.image-output {
    width: 100%;
    max-width: 800px;
    margin: 0 auto;
    display: block;
}

.prompt-box {
    width: 90%;
    max-width: 1000px;
    margin: 20px auto;
}

.generate-btn {
    background: linear-gradient(45deg, #ff6b6b, #ff8e8e) !important;
    color: white !important;
    padding: 15px 30px !important;
    border-radius: 8px !important;
    border: none !important;
    font-size: 1.1em !important;
    cursor: pointer !important;
    transition: all 0.3s ease !important;
    width: 250px !important;
    margin: 20px auto !important;
    display: block !important;
    text-transform: uppercase !important;
    letter-spacing: 1px !important;
}

.generate-btn:hover {
    background: linear-gradient(45deg, #ff8e8e, #ff6b6b) !important;
    transform: translateY(-2px) !important;
    box-shadow: 0 5px 15px rgba(255, 107, 107, 0.4) !important;
}

.accordion {
    width: 90%;
    max-width: 1000px;
    margin: 20px auto;
}

.examples-table {
    background-color: rgba(255, 255, 255, 0.95);
    border-radius: 8px;
    padding: 15px;
    margin-top: 20px;
}

.parameter-box {
    background-color: #f8f9fa;
    padding: 20px;
    border-radius: 8px;
    margin: 10px 0;
}
"""


with gr.Blocks(theme="Yntec/HaleyCH_Theme_Orange") as app:
    gr.Markdown("# ๐ŸŽญ Fashion AI Studio")
    gr.Markdown("Generate fashion images and try on virtual clothing using AI")
    
    with gr.Tabs():
        # Virtual Try-On ํƒญ
        with gr.TabItem("๐Ÿ‘” Virtual Try-On"):
            with gr.Row():
                with gr.Column():
                    prompt_input = gr.Textbox(
                        label="Style Description", 
                        placeholder="Describe the desired style (e.g., 'person wearing elegant dress')"
                    )
                    with gr.Row():
                        with gr.Group():
                            structure_image = gr.Image(
                                label="Your Photo (Full-body)", 
                                type="filepath"
                            )
                            gr.Markdown("*Upload a clear, well-lit full-body photo*")
                            depth_strength = gr.Slider(
                                minimum=0, 
                                maximum=50, 
                                value=15, 
                                label="Fitting Strength"
                            )
                        with gr.Group():
                            style_image = gr.Image(
                                label="Clothing Item", 
                                type="filepath"
                            )
                            gr.Markdown("*Upload the clothing item you want to try on*")
                            style_strength = gr.Slider(
                                minimum=0, 
                                maximum=1, 
                                value=0.5, 
                                label="Style Transfer Strength"
                            )
                    tryon_btn = gr.Button("Generate Try-On")
                    
                    gr.Examples(
                        examples=examples,
                        inputs=[prompt_input, structure_image, style_image, depth_strength, style_strength],
                        outputs=[output_image],
                        fn=generate_image,
                        cache_examples=True
                    )
                
                with gr.Column():
                    output_image.render()

        # Fashion Generation ํƒญ
        with gr.TabItem("๐Ÿ‘— Fashion Generation"):
            with gr.Column():
                # ๋ชจ๋“œ ์„ ํƒ
                with gr.Group():
                    mode = gr.Radio(
                        choices=["Generate Model", "Generate Clothes"],
                        label="Generation Mode",
                        value="Generate Model"
                    )
                
                # ํ”„๋กฌํ”„ํŠธ ์ž…๋ ฅ
                prompt = gr.TextArea(
                    label="โœ๏ธ Fashion Description (ํ•œ๊ธ€ ๋˜๋Š” ์˜์–ด)",
                    placeholder="ํŒจ์…˜ ๋ชจ๋ธ์ด๋‚˜ ์˜๋ฅ˜๋ฅผ ์„ค๋ช…ํ•˜์„ธ์š”...",
                    lines=5
                )
                
                # ๊ฒฐ๊ณผ ์ด๋ฏธ์ง€
                result = gr.Image(label="Generated Fashion")
                
                generate_button = gr.Button("๐Ÿš€ Generate Fashion")
                
                # ๊ณ ๊ธ‰ ์„ค์ • ์•„์ฝ”๋””์–ธ
                with gr.Accordion("๐ŸŽจ Advanced Options", open=False):
                    with gr.Row():
                        cfg_scale = gr.Slider(label="CFG Scale", minimum=1, maximum=20, value=7.0)
                        steps = gr.Slider(label="Steps", minimum=1, maximum=100, value=30)
                        lora_scale = gr.Slider(label="LoRA Scale", minimum=0, maximum=1, value=0.85)
                    
                    with gr.Row():
                        width = gr.Slider(label="Width", minimum=256, maximum=1536, value=512)
                        height = gr.Slider(label="Height", minimum=256, maximum=1536, value=768)
                    
                    with gr.Row():
                        randomize_seed = gr.Checkbox(True, label="Randomize seed")
                        seed = gr.Slider(label="Seed", minimum=0, maximum=MAX_SEED, value=42)

                # ์˜ˆ์ œ ํƒญ
                with gr.Tabs():
                    with gr.TabItem("Model Examples"):
                        gr.Examples(examples=example_model_prompts, inputs=prompt)
                    with gr.TabItem("Clothes Examples"):
                        gr.Examples(examples=example_clothes_prompts, inputs=prompt)

    # ์ด๋ฒคํŠธ ํ•ธ๋“ค๋Ÿฌ
    tryon_btn.click(
        fn=generate_image,
        inputs=[prompt_input, structure_image, style_image, depth_strength, style_strength],
        outputs=[output_image]
    )
    
    generate_button.click(
        generate_fashion,
        inputs=[prompt, mode, cfg_scale, steps, randomize_seed, seed, width, height, lora_scale],
        outputs=[result, seed]
    )

if __name__ == "__main__":
    app.launch(share=True)