Add README.md

README.md

@@ -1,15 +1,44 @@
 ---
-sdk: gradio
 title: Comic Character Extractor
-emoji: 🐨
-colorFrom: purple
-colorTo: purple
-sdk_version: 6.24.0
-python_version: '3.13'
+emoji: 🎨
+colorFrom: blue
+colorTo: blue
+sdk: gradio
+sdk_version: 6.22.0
 app_file: app.py
-pinned: false
-tags:
-- ml-intern
+pinned: true
+license: apache-2.0
+short_description: Auto-extract characters from comic pages
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+## Comic Character Extractor
+
+Upload a comic page image and the system **automatically detects, extracts, and re-renders every character** on the page using a neutral reference prompt (plain background, no emotion, no pose, no action).
+
+### How It Works
+
+1. **Upload** a comic page image
+2. **Grounding DINO (tiny)** detects all characters/people in the page
+3. **SAM (Segment Anything)** segments each character with precise masks
+4. Each character is cropped onto a **plain white background**
+5. **Qwen-Image-Edit-2511** re-renders each with a **neutral prompt**:
+
+> "Extracted character reference image, front view, standing upright, neutral facial expression, no emotion, plain solid background, no action, no pose, studio lighting, clean white background"
+
+### Features
+
+- **Automatic character count** — detects as many characters as are on the page
+- **Neutral reference prompts** — no emotion, no action, no pose, plain background
+- **Batch extraction** — all characters extracted in one go
+
+### Models Used
+- **Detector**: [IDEA-Research/grounding-dino-tiny](https://huggingface.co/IDEA-Research/grounding-dino-tiny)
+- **Segmenter**: [facebook/sam-vit-base](https://huggingface.co/facebook/sam-vit-base)
+- **Editor**: [Qwen/Qwen-Image-Edit-2511](https://huggingface.co/Qwen/Qwen-Image-Edit-2511)
+- **Base Space**: Inspired by [prithivMLmods/Qwen-Image-Edit-2511-LoRAs-Fast](https://huggingface.co/spaces/prithivMLmods/Qwen-Image-Edit-2511-LoRAs-Fast)
+
+### Usage
+1. Upload a comic page image (drag & drop or paste with Ctrl+V)
+2. Click "Run" or press Ctrl+Enter
+3. Wait for automatic detection & extraction
+4. Browse extracted characters in the filmstrip