Spaces:
Sleeping
Sleeping
|
Download README.md from EnYa32/FacialKeypointDetectionCNN: direct link, hf CLI and curl.
- Browser
- Download file 1.14 kB
-
https://huggingface.co/spaces/EnYa32/FacialKeypointDetectionCNN/resolve/main/README.md
- Command line
-
hf download hf://spaces/EnYa32/FacialKeypointDetectionCNN/README.md
-
curl -L -o README.md https://huggingface.co/spaces/EnYa32/FacialKeypointDetectionCNN/resolve/main/README.md
1.14 kB
A newer version of the Streamlit SDK is available: 1.67.0
metadata
title: FacialKeypointDetectionCNN
emoji: π
colorFrom: red
colorTo: red
sdk: streamlit
app_file: src/streamlit_app.py
pinned: false
short_description: 'Predict facial keypoints from an uploaded face image '
license: mit
π Facial Keypoints Detection (CNN)
This app predicts facial keypoints (eyes, eyebrows, nose, mouth) from a face image.
What this project does
- Input: face image (grayscale or RGB)
- Output: 30 values (x/y coordinates for 15 facial keypoints)
- The app overlays keypoints on a 96Γ96 image and shows the coordinates in a table.
Files in this repo
app.pyβ Streamlit appfinal_keypoints_cnn.kerasβ trained Keras modeltarget_cols.jsonβ output column names (order of the 30 targets)preprocess_config.jsonβ preprocessing settings (image size, normalization)
How to run locally
pip install -r requirements.txt
streamlit run app.py
Preprocessing (same as training)
Convert to grayscale
Resize to 96Γ96
Normalize pixels: x / 255.0
Model predicts normalized coordinates
Convert back to pixel space: y = y * 48 + 48
Clip to valid range: [0, 96]