π AI Emotion Detector
Real-time facial emotion detection using state-of-the-art Vision Transformer!
π Features
- πΉ Live Webcam Detection - Capture and analyze emotions in real-time
- πΌοΈ Image Upload - Drag & drop or upload any face image
- π― High Accuracy - 98.80% validation accuracy
- π¨ Beautiful UI - Animated results with confidence scores
- 7οΈβ£ 7 Emotions - Angry, Disgust, Fear, Happy, Sad, Surprise, Neutral
π Model Details
- Model ID:
koyelog/face - Architecture: Vision Transformer (ViT)
- Parameters: 85.8 Million
- Training Dataset: 181,230 images
- Training Accuracy: 99.29%
- Validation Accuracy: 98.80%
- License: MIT
π Supported Emotions
| Emotion | Emoji | Color |
|---|---|---|
| Angry | π | Red |
| Disgust | π€’ | Green |
| Fear | π¨ | Purple |
| Happy | π | Yellow |
| Sad | π’ | Blue |
| Surprise | π² | Magenta |
| Neutral | π | Gray |
π‘ Usage Tips
For best results:
- β Use front-facing photos
- β Ensure good lighting
- β Single face per image
- β Clear facial expressions
π©βπ» Created By
Koyeliya Ghosh
- HuggingFace: @koyelog
- Model: koyelog/face
π License
MIT License - Free to use and modify!
- Downloads last month
- 177
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
π
Ask for provider support