Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,70 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
license: mit
|
| 4 |
+
tags:
|
| 5 |
+
- asl
|
| 6 |
+
- sign-language
|
| 7 |
+
- cnn
|
| 8 |
+
- tensorflow
|
| 9 |
+
- mediapipe
|
| 10 |
+
- computer-vision
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# 🧠 ASL Sign Language Detection with CNN (Conv1D)
|
| 14 |
+
|
| 15 |
+
Model ini dikembangkan untuk mendeteksi huruf American Sign Language (ASL) dari koordinat tangan (landmark) secara real-time menggunakan CNN (Convolutional Neural Network).
|
| 16 |
+
|
| 17 |
+
## 📌 Deskripsi
|
| 18 |
+
|
| 19 |
+
- Input: 63 fitur (x, y, z) dari 21 titik tangan yang diekstraksi dengan [MediaPipe Hands](https://google.github.io/mediapipe/solutions/hands.html)
|
| 20 |
+
- Model: CNN dengan `Conv1D`, dilatih menggunakan TensorFlow
|
| 21 |
+
- Dataset: ASL Alphabet (87.000+ gambar, 29 kelas)
|
| 22 |
+
|
| 23 |
+
## 🏋️ Training
|
| 24 |
+
|
| 25 |
+
```
|
| 26 |
+
Epoch 27/50
|
| 27 |
+
accuracy: 98.80% - val_accuracy: 98.78% - loss: 0.0386 - val_loss: 0.0636
|
| 28 |
+
```
|
| 29 |
+
|
| 30 |
+
## 📊 Evaluasi Model
|
| 31 |
+
|
| 32 |
+
- Akurasi Uji: **98.92%**
|
| 33 |
+
- Total dataset : 63.676 sampel
|
| 34 |
+
- Jumlah data latih : 50.940 sampel
|
| 35 |
+
- Jumlah data uji : 12.736 sampel
|
| 36 |
+
|
| 37 |
+
### 📋 Classification Report
|
| 38 |
+
|
| 39 |
+
| Label | Precision | Recall | F1-Score | Support |
|
| 40 |
+
|-----------|-----------|--------|----------|---------|
|
| 41 |
+
| A | 0.98 | 1.00 | 0.99 | 437 |
|
| 42 |
+
| B | 1.00 | 1.00 | 1.00 | 441 |
|
| 43 |
+
| ... | ... | ... | ... | ... |
|
| 44 |
+
| Z | 0.99 | 0.99 | 0.99 | 470 |
|
| 45 |
+
| SPACE | 0.98 | 0.99 | 0.99 | 325 |
|
| 46 |
+
| **Average** | **0.95** | **0.95** | **0.95** | **12736** |
|
| 47 |
+
|
| 48 |
+
> *Note: Label "NOTHING" memiliki support sangat kecil (1 sampel), sehingga metriknya tidak signifikan.*
|
| 49 |
+
|
| 50 |
+
## 📦 Dependencies
|
| 51 |
+
|
| 52 |
+
- TensorFlow
|
| 53 |
+
- MediaPipe
|
| 54 |
+
- scikit-learn
|
| 55 |
+
- NumPy
|
| 56 |
+
|
| 57 |
+
## 📹 Demo
|
| 58 |
+
|
| 59 |
+
Model ini bisa digunakan untuk prediksi real-time langsung dari webcam di browser setelah dikonversi ke TensorFlow.js.
|
| 60 |
+
|
| 61 |
+
## 👤 Penulis
|
| 62 |
+
|
| 63 |
+
- Ade Maulana
|
| 64 |
+
- IG: [@ademaulana_](https://instagram.com/ademaulana_)
|
| 65 |
+
- TikTok: [@ademaulana_4](https://tiktok.com/@ademaulana_4)
|
| 66 |
+
|
| 67 |
+
## 📄 License
|
| 68 |
+
|
| 69 |
+
MIT License
|
| 70 |
+
|