Update README.md

README.md

@@ -1,6 +1,6 @@
 ---
 title: Korean Object Detection
-emoji: 🦀
+emoji: 👀
 colorFrom: yellow
 colorTo: yellow
 sdk: static
@@ -10,3 +10,69 @@ short_description: Real-time object detection using COCO-SSD in the browser
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+# Korean Object Detection App (COCO-SSD)
+
+This project is a real-time object detection web app that overlays Korean vocabulary labels on top of detected objects using the **COCO-SSD model (TensorFlow.js)**. It’s built for both AI experimentation and Korean language learning — with mobile-first optimization and Docker deployment.
+
+---
+
+## 🧠 Why This is Awesome
+
+This isn’t just an object detector — it’s a **language learning tool**.
+
+You point your camera at real objects — a cup, a dog, a book — and it teaches you the Korean word for each one in real time. Think of it like flashcards... but in your actual house.
+
+Perfect for:
+
+- Korean learners
+- Tourists in Seoul
+- Kids growing up abroad
+- Anyone who hates memorizing vocab lists
+
+---
+
+## 🌐 Tech Stack
+
+- **COCO-SSD** — real-time object detection via TensorFlow.js
+- **HTML Canvas** — draws bounding boxes and Korean labels
+
+---
+
+## 📱 Mobile Optimization
+
+- Resolution capped at 640×480
+- Inference runs every ~300ms
+- Lightweight canvas redraw
+- Runs well on iOS Safari (maybe a little slow)
+
+---
+
+## 🇰🇷 Korean Vocabulary Mapping
+
+All 80 COCO-SSD object classes are labeled in Korean (with fallback to English). Example:
+
+```json
+{
+  "dog": "강아지",
+  "person": "사람",
+  "book": "책",
+  "cell phone": "휴대폰"
+}
+```
+
+---
+
+## ✨ Future Features
+
+- TOPIK level filtering
+- Audio pronunciation (TTS)
+- Vocabulary challenges
+- Voice-based guessing game
+- “Kid Mode” with points & stickers
+
+---
+
+## 🛠 Maintainer
+
+Made with frustration, triumph, and lots of 사랑 by Ramsi K.