Add Korean localized Qwen Image Editor with LFS

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.jpg filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text

CLAUDE.md

@@ -0,0 +1,43 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Architecture Overview
+
+This is a Hugging Face Spaces application that provides image editing capabilities using the Qwen-Image-Edit model. The application consists of a single Gradio interface (`app.py`) that orchestrates:
+
+1. **Image Processing Pipeline**: Uses `QwenImageEditPipeline` from diffusers for the core image editing functionality
+2. **Prompt Enhancement**: Integrates with DashScope API (`qwen-vl-max-latest`) to automatically rewrite and enhance user edit instructions using a detailed system prompt
+3. **Gradio Interface**: Web UI with image upload, text input, and advanced parameter controls
+
+## Key Components
+
+- **Main Pipeline**: `QwenImageEditPipeline` loaded from "Qwen/Qwen-Image-Edit" model
+- **Prompt Polishing**: `polish_prompt()` function that uses DashScope API to enhance user instructions via a comprehensive system prompt template
+- **Inference Function**: `infer()` decorated with `@spaces.GPU(duration=120)` for GPU acceleration
+- **UI Layout**: Single-page Gradio interface with examples and advanced settings accordion
+
+## Environment Requirements
+
+- **GPU**: CUDA-enabled environment preferred (falls back to CPU)
+- **API Key**: `DASH_API_KEY` environment variable required for prompt enhancement via DashScope
+- **Dependencies**: PyTorch, diffusers (git version), transformers, dashscope, gradio
+
+## Development Commands
+
+Install dependencies:
+```bash
+pip install -r requirements.txt
+```
+
+Run the application:
+```bash
+python app.py
+```
+
+## Configuration Notes
+
+- Default inference settings: 50 steps, guidance scale 4.0, bfloat16 dtype
+- Hardcoded negative prompt: single space character
+- GPU duration limit: 120 seconds for Spaces environment
+- Examples include three preset image/prompt combinations for testing

README.md

@@ -1,12 +1,62 @@
 ---
-title: Qwen Image Edit
-emoji: 😻
-colorFrom: indigo
+title: 퀀 이미지 편집기
+emoji: ✒️
+colorFrom: pink
 colorTo: green
 sdk: gradio
 sdk_version: 5.43.1
 app_file: app.py
 pinned: false
+license: apache-2.0
+python_version: 3.11
+---
+
+# 퀀 이미지 편집기 (Qwen Image Editor)
+
+이 앱은 Qwen-Image-Edit 모델을 사용한 한국어 이미지 편집기입니다.
+
+## 한국어 로컬라이제이션 변경사항
+
+이 프로젝트는 완전히 한국어로 로컬라이제이션되었습니다:
+
+### UI 컴포넌트
+- **제목**: "퀀 이미지 편집기" (로고 + 텍스트)
+- **이미지 업로드 영역**: "이미지를 여기로 끌어다 놓으세요 - 또는 - 업로드하려면 클릭하세요"
+- **프롬프트 입력**: "편집 지시사항을 설명해주세요" (플레이스홀더)
+- **버튼**: "편집!" (실행 버튼)
+- **고급 설정 섹션**: "고급 설정"
+
+### 설정 옵션
+- **시드**: "시드" 및 "시드 랜덤화"
+- **가이던스 스케일**: "가이던스 스케일"
+- **추론 단계**: "추론 단계 수"
+- **프롬프트 재작성**: "프롬프트 재작성"
+
+### 예시 섹션
+- **섹션 제목**: "예시"
+- **예시 프롬프트들** (UI에는 한국어 표시, 내부적으로는 영어로 변환하여 처리):
+  1. "텍스트를 'COOL NEON SIGN HERE'으로 변경해주세요" → "Change the text to 'COOL NEON SIGN HERE'"
+  2. "고양이가 검은색 전통 갓을 쓰고 있는 모습으로 만들어주세요" → "A cat wearing a traditional Korean gat (black wide-brimmed horsehair hat)"
+  3. "사진 스타일을 빈티지 만화책 스타일로 바꿔주세요" → "Change the photo style to vintage comic book"
+
+### 기술적 변경사항
+- **로고**: Hugging Face resolve URL을 사용하여 로컬 logo.png 파일 참조
+- **정적 파일 서빙**: `gr.set_static_paths()` 추가
+- **설명 텍스트**: GitHub 링크 및 사용 안내를 한국어로 번역
+- **GPU 타임아웃**: 180초(3분)로 설정하여 안정적인 처리 보장
+- **에러 핸들링**: 예외 발생 시 원본 이미지 반환 및 재시도 로직 개선
+- **다운로드 형식**: PNG 형식으로 설정하여 더 나은 이미지 품질 제공
+- **프롬프트 재작성**: API 호출 실패 시 최대 3회 재시도 후 원본 프롬프트 사용
+- **시드 랜덤화**: 기본값으로 활성화되어 매번 다른 결과 생성, 일관된 UI/함수 동작 보장
+- **이중 언어 지원**: 한국어 UI 표시 + 내부적 영어 번역으로 사용자 친화성과 AI 성능을 동시에 최적화
+
+## 원본 모델 정보
+
+자세한 내용은 [Qwen-Image 시리즈](https://github.com/QwenLM/Qwen-Image)를 참조하세요. 
+[Qwen Chat](https://chat.qwen.ai/)에서 체험하거나 [모델을 다운로드](https://huggingface.co/Qwen/Qwen-Image-Edit)하여 ComfyUI나 diffusers로 로컬에서 실행할 수 있습니다.
+
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+<!-- Force restart -->

app.py

@@ -0,0 +1,320 @@
+import gradio as gr
+import numpy as np
+import random
+import torch
+import spaces
+
+from PIL import Image
+from diffusers import QwenImageEditPipeline
+
+import os
+import base64
+import json
+
+# Set static paths for serving logo
+gr.set_static_paths(paths=["./"])
+
+
+SYSTEM_PROMPT = '''
+# Edit Instruction Rewriter
+You are a professional edit instruction rewriter. Your task is to generate a precise, concise, and visually achievable professional-level edit instruction based on the user-provided instruction and the image to be edited.  
+
+Please strictly follow the rewriting rules below:
+
+## 1. General Principles
+- Keep the rewritten prompt **concise**. Avoid overly long sentences and reduce unnecessary descriptive language.  
+- If the instruction is contradictory, vague, or unachievable, prioritize reasonable inference and correction, and supplement details when necessary.  
+- Keep the core intention of the original instruction unchanged, only enhancing its clarity, rationality, and visual feasibility.  
+- All added objects or modifications must align with the logic and style of the edited input image’s overall scene.  
+
+## 2. Task Type Handling Rules
+### 1. Add, Delete, Replace Tasks
+- If the instruction is clear (already includes task type, target entity, position, quantity, attributes), preserve the original intent and only refine the grammar.  
+- If the description is vague, supplement with minimal but sufficient details (category, color, size, orientation, position, etc.). For example:  
+    > Original: "Add an animal"  
+    > Rewritten: "Add a light-gray cat in the bottom-right corner, sitting and facing the camera"  
+- Remove meaningless instructions: e.g., "Add 0 objects" should be ignored or flagged as invalid.  
+- For replacement tasks, specify "Replace Y with X" and briefly describe the key visual features of X.  
+
+### 2. Text Editing Tasks
+- All text content must be enclosed in English double quotes `" "`. Do not translate or alter the original language of the text, and do not change the capitalization.  
+- **For text replacement tasks, always use the fixed template:**
+    - `Replace "xx" to "yy"`.  
+    - `Replace the xx bounding box to "yy"`.  
+- If the user does not specify text content, infer and add concise text based on the instruction and the input image’s context. For example:  
+    > Original: "Add a line of text" (poster)  
+    > Rewritten: "Add text \"LIMITED EDITION\" at the top center with slight shadow"  
+- Specify text position, color, and layout in a concise way.  
+
+### 3. Human Editing Tasks
+- Maintain the person’s core visual consistency (ethnicity, gender, age, hairstyle, expression, outfit, etc.).  
+- If modifying appearance (e.g., clothes, hairstyle), ensure the new element is consistent with the original style.  
+- **For expression changes, they must be natural and subtle, never exaggerated.**  
+- If deletion is not specifically emphasized, the most important subject in the original image (e.g., a person, an animal) should be preserved.
+    - For background change tasks, emphasize maintaining subject consistency at first.  
+- Example:  
+    > Original: "Change the person’s hat"  
+    > Rewritten: "Replace the man’s hat with a dark brown beret; keep smile, short hair, and gray jacket unchanged"  
+
+### 4. Style Transformation or Enhancement Tasks
+- If a style is specified, describe it concisely with key visual traits. For example:  
+    > Original: "Disco style"  
+    > Rewritten: "1970s disco: flashing lights, disco ball, mirrored walls, colorful tones"  
+- If the instruction says "use reference style" or "keep current style," analyze the input image, extract main features (color, composition, texture, lighting, art style), and integrate them concisely.  
+- **For coloring tasks, including restoring old photos, always use the fixed template:** "Restore old photograph, remove scratches, reduce noise, enhance details, high resolution, realistic, natural skin tones, clear facial features, no distortion, vintage photo restoration"  
+- If there are other changes, place the style description at the end.
+
+## 3. Rationality and Logic Checks
+- Resolve contradictory instructions: e.g., "Remove all trees but keep all trees" should be logically corrected.  
+- Add missing key information: if position is unspecified, choose a reasonable area based on composition (near subject, empty space, center/edges).  
+
+# Output Format Example
+```json
+{
+   "Rewritten": "..."
+}
+'''
+
+def polish_prompt(prompt, img):
+    original_prompt = prompt
+    prompt = f"{SYSTEM_PROMPT}\n\nUser Input: {prompt}\n\nRewritten Prompt:"
+    success=False
+    max_retries = 3
+    retry_count = 0
+    
+    while not success and retry_count < max_retries:
+        try:
+            result = api(prompt, [img])
+            # print(f"Result: {result}")
+            # print(f"Polished Prompt: {polished_prompt}")
+            if isinstance(result, str):
+                result = result.replace('```json','')
+                result = result.replace('```','')
+                result = json.loads(result)
+            else:
+                result = json.loads(result)
+
+            polished_prompt = result['Rewritten']
+            polished_prompt = polished_prompt.strip()
+            polished_prompt = polished_prompt.replace("\n", " ")
+            success = True
+        except Exception as e:
+            print(f"[Warning] Error during API call (attempt {retry_count + 1}): {e}")
+            retry_count += 1
+            
+    if not success:
+        print(f"[Warning] Failed to polish prompt after {max_retries} attempts, using original prompt")
+        return original_prompt
+        
+    return polished_prompt
+
+
+def encode_image(pil_image):
+    import io
+    buffered = io.BytesIO()
+    pil_image.save(buffered, format="PNG")
+    return base64.b64encode(buffered.getvalue()).decode("utf-8")
+
+
+
+
+def api(prompt, img_list, model="qwen-vl-max-latest", kwargs={}):
+    import dashscope
+    api_key = os.environ.get('DASH_API_KEY')
+    if not api_key:
+        raise EnvironmentError("DASH_API_KEY is not set")
+    assert model in ["qwen-vl-max-latest"], f"Not implemented model {model}"
+    sys_promot = "you are a helpful assistant, you should provide useful answers to users."
+    messages = [
+        {"role": "system", "content": sys_promot},
+        {"role": "user", "content": []}]
+    for img in img_list:
+        messages[1]["content"].append(
+            {"image": f"data:image/png;base64,{encode_image(img)}"})
+    messages[1]["content"].append({"text": f"{prompt}"})
+
+    response_format = kwargs.get('response_format', None)
+
+    response = dashscope.MultiModalConversation.call(
+        api_key=api_key,
+        model=model, # For example, use qwen-plus here. You can change the model name as needed. Model list: https://help.aliyun.com/zh/model-studio/getting-started/models
+        messages=messages,
+        result_format='message',
+        response_format=response_format,
+        )
+
+    if response.status_code == 200:
+        return response.output.choices[0].message.content[0]['text']
+    else:
+        raise Exception(f'Failed to post: {response}')
+
+# --- Model Loading ---
+dtype = torch.bfloat16
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+# Load the model pipeline
+pipe = QwenImageEditPipeline.from_pretrained("Qwen/Qwen-Image-Edit", torch_dtype=dtype).to(device)
+
+# --- UI Constants and Helpers ---
+MAX_SEED = np.iinfo(np.int32).max
+
+# --- Main Inference Function (with hardcoded negative prompt) ---
+@spaces.GPU(duration=180)
+def infer(
+    image,
+    prompt,
+    seed=0,
+    randomize_seed=True,
+    true_guidance_scale=1.0,
+    num_inference_steps=50,
+    rewrite_prompt=True,
+    progress=gr.Progress(track_tqdm=True),
+):
+    """
+    Generates an image using the local Qwen-Image diffusers pipeline.
+    """
+    # Hardcode the negative prompt as requested
+    negative_prompt = " "
+    
+    if randomize_seed:
+        seed = random.randint(0, MAX_SEED)
+
+    # Set up the generator for reproducibility
+    generator = torch.Generator(device=device).manual_seed(seed)
+    
+    # Translate Korean example prompts to English for better AI performance
+    korean_to_english_examples = {
+        "텍스트를 'COOL NEON SIGN HERE'으로 변경해주세요": "Change the text to 'COOL NEON SIGN HERE'",
+        "고양이가 검은색 전통 갓을 쓰고 있는 모습으로 만들어주세요": "A cat wearing a traditional Korean gat (black wide-brimmed horsehair hat)",
+        "사진 스타일을 빈티지 만화책 스타일로 바꿔주세요": "Change the photo style to vintage comic book"
+    }
+    
+    # Use English version if this is one of our Korean examples
+    if prompt in korean_to_english_examples:
+        original_prompt = prompt
+        prompt = korean_to_english_examples[prompt]
+        print(f"Translated Korean example: '{original_prompt}' -> '{prompt}'")
+    
+    print(f"Calling pipeline with prompt: '{prompt}'")
+    print(f"Negative Prompt: '{negative_prompt}'")
+    print(f"Seed: {seed}, Steps: {num_inference_steps}, Guidance: {true_guidance_scale}")
+    try:
+        if rewrite_prompt:
+            prompt = polish_prompt(prompt, image)
+            print(f"Rewritten Prompt: {prompt}")
+
+        # Generate the image
+        images = pipe(
+            image,
+            prompt=prompt,
+            negative_prompt=negative_prompt,
+            num_inference_steps=num_inference_steps,
+            generator=generator,
+            true_cfg_scale=true_guidance_scale,
+            num_images_per_prompt=1
+        ).images
+        
+        return images[0], seed
+    
+    except Exception as e:
+        print(f"Error during inference: {e}")
+        # Return the original image with error message
+        return image, seed
+
+# --- Examples and UI Layout ---
+examples = []
+
+css = """
+#col-container {
+    margin: 0 auto;
+    max-width: 1024px;
+}
+#edit_text{
+    margin-top: -62px !important
+}
+"""
+
+with gr.Blocks(css=css) as demo:
+    with gr.Column(elem_id="col-container"):
+        gr.HTML('<h1 style="text-align: center; color: #6366f1; font-size: 3rem; font-weight: bold; margin: 2rem 0; font-family: system-ui, -apple-system, sans-serif; display: flex; align-items: center; justify-content: center; gap: 1rem;"><img src="https://huggingface.co/spaces/tchung1970/Qwen-Image-Edit/resolve/main/logo.png" alt="로고" style="height: 3rem; width: auto;"> 퀀 이미지 편집기</h1>')
+        gr.Markdown("[더 알아보기](https://github.com/QwenLM/Qwen-Image)에서 Qwen-Image 시리즈에 대해 자세히 알아보세요. [Qwen Chat](https://chat.qwen.ai/)에서 체험하거나 [모델 다운로드](https://huggingface.co/Qwen/Qwen-Image-Edit)하여 ComfyUI나 diffusers로 로컬에서 실행해보세요.")
+        with gr.Row():
+            with gr.Column():
+                input_image = gr.Image(
+                    label="입력 이미지", 
+                    show_label=False, 
+                    type="pil",
+                    placeholder="이미지를 여기로 끌어다 놓으세요\n- 또는 -\n업로드하려면 클릭하세요"
+                )
+
+            result = gr.Image(label="결과", show_label=False, type="pil", format="png")
+        with gr.Row():
+            prompt = gr.Text(
+                    label="프롬프트",
+                    show_label=False,
+                    placeholder="편집 지시사항을 설명해주세요",
+                    container=False,
+            )
+            run_button = gr.Button("편집!", variant="primary")
+
+        with gr.Accordion("고급 설정", open=False):
+            # Negative prompt UI element is removed here
+
+            seed = gr.Slider(
+                label="시드",
+                minimum=0,
+                maximum=MAX_SEED,
+                step=1,
+                value=0,
+            )
+
+            randomize_seed = gr.Checkbox(label="시드 랜덤화", value=True)
+
+            with gr.Row():
+
+                true_guidance_scale = gr.Slider(
+                    label="가이던스 스케일",
+                    minimum=1.0,
+                    maximum=10.0,
+                    step=0.1,
+                    value=4.0
+                )
+
+                num_inference_steps = gr.Slider(
+                    label="추론 단계 수",
+                    minimum=1,
+                    maximum=50,
+                    step=1,
+                    value=50,
+                )
+                
+                rewrite_prompt = gr.Checkbox(label="프롬프트 재작성", value=True)
+
+        gr.Examples(
+                label="예시",
+                examples=[
+                ["neon_sign.png", "텍스트를 'COOL NEON SIGN HERE'으로 변경해주세요"],
+                ["cat_sitting.jpg", "고양이가 검은색 전통 갓을 쓰고 있는 모습으로 만들어주세요"],
+                ["pie.png", "사진 스타일을 빈티지 만화책 스타일로 바꿔주세요"]],
+                    inputs=[input_image, prompt], 
+                    outputs=[result, seed], 
+                    fn=infer, 
+                    cache_examples="lazy")
+
+    gr.on(
+        triggers=[run_button.click, prompt.submit],
+        fn=infer,
+        inputs=[
+            input_image,
+            prompt,
+            seed,
+            randomize_seed,
+            true_guidance_scale,
+            num_inference_steps,
+            rewrite_prompt,
+        ],
+        outputs=[result, seed],
+    )
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,9 @@
+# Force rebuild
+git+https://github.com/huggingface/diffusers.git
+
+transformers
+accelerate
+safetensors
+sentencepiece
+dashscope
+torchvision