mirror of
https://github.com/yakhyo/uniface.git
synced 2026-05-15 12:57:55 +00:00
Compare commits
2 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
426bd71505 | ||
|
|
ede8b27091 |
2
.github/workflows/ci.yml
vendored
2
.github/workflows/ci.yml
vendored
@@ -33,6 +33,8 @@ jobs:
|
||||
matrix:
|
||||
include:
|
||||
# Full Python range on Linux (fastest runner)
|
||||
- os: ubuntu-latest
|
||||
python-version: "3.10"
|
||||
- os: ubuntu-latest
|
||||
python-version: "3.11"
|
||||
- os: ubuntu-latest
|
||||
|
||||
2
.github/workflows/publish.yml
vendored
2
.github/workflows/publish.yml
vendored
@@ -54,7 +54,7 @@ jobs:
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
python-version: ["3.11", "3.13"]
|
||||
python-version: ["3.10", "3.11", "3.13"]
|
||||
|
||||
steps:
|
||||
- name: Checkout code
|
||||
|
||||
111
README.md
111
README.md
@@ -3,7 +3,7 @@
|
||||
<div align="center">
|
||||
|
||||
[](https://pypi.org/project/uniface/)
|
||||
[](https://www.python.org/)
|
||||
[](https://www.python.org/)
|
||||
[](https://opensource.org/licenses/MIT)
|
||||
[](https://github.com/yakhyo/uniface/actions)
|
||||
[](https://pepy.tech/projects/uniface)
|
||||
@@ -33,7 +33,7 @@
|
||||
- **Gaze Estimation** — Real-time gaze direction with MobileGaze
|
||||
- **Head Pose Estimation** — 3D head orientation (pitch, yaw, roll) with 6D rotation representation
|
||||
- **Attribute Analysis** — Age, gender, race (FairFace), and emotion
|
||||
- **Vector Indexing** — FAISS-backed embedding store for fast multi-identity search
|
||||
- **Vector Store** — FAISS-backed embedding store for fast multi-identity search
|
||||
- **Anti-Spoofing** — Face liveness detection with MiniFASNet
|
||||
- **Face Anonymization** — 5 blur methods for privacy protection
|
||||
- **Hardware Acceleration** — ARM64 (Apple Silicon), CUDA (NVIDIA), CPU
|
||||
@@ -61,7 +61,7 @@ git clone https://github.com/yakhyo/uniface.git
|
||||
cd uniface && pip install -e .
|
||||
```
|
||||
|
||||
**FAISS vector indexing**
|
||||
**FAISS vector store**
|
||||
|
||||
```bash
|
||||
pip install faiss-cpu # or faiss-gpu for CUDA
|
||||
@@ -127,14 +127,10 @@ for face in faces:
|
||||
|
||||
```python
|
||||
import cv2
|
||||
from uniface.analyzer import FaceAnalyzer
|
||||
from uniface.detection import RetinaFace
|
||||
from uniface.recognition import ArcFace
|
||||
from uniface import FaceAnalyzer
|
||||
|
||||
detector = RetinaFace()
|
||||
recognizer = ArcFace()
|
||||
|
||||
analyzer = FaceAnalyzer(detector, recognizer=recognizer)
|
||||
# Zero-config: uses SCRFD (500M) + ArcFace (MobileNet) by default
|
||||
analyzer = FaceAnalyzer()
|
||||
|
||||
image = cv2.imread("photo.jpg")
|
||||
if image is None:
|
||||
@@ -146,52 +142,18 @@ for face in faces:
|
||||
print(face.bbox, face.embedding.shape if face.embedding is not None else None)
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Execution Providers (ONNX Runtime)
|
||||
With attributes:
|
||||
|
||||
```python
|
||||
from uniface.detection import RetinaFace
|
||||
from uniface import FaceAnalyzer, AgeGender
|
||||
|
||||
# Force CPU-only inference
|
||||
detector = RetinaFace(providers=["CPUExecutionProvider"])
|
||||
analyzer = FaceAnalyzer(attributes=[AgeGender()])
|
||||
faces = analyzer.analyze(image)
|
||||
|
||||
for face in faces:
|
||||
print(f"{face.sex}, {face.age}y, embedding={face.embedding.shape}")
|
||||
```
|
||||
|
||||
See more in the docs:
|
||||
https://yakhyo.github.io/uniface/concepts/execution-providers/
|
||||
|
||||
---
|
||||
|
||||
## Documentation
|
||||
|
||||
Full documentation: https://yakhyo.github.io/uniface/
|
||||
|
||||
| Resource | Description |
|
||||
|----------|-------------|
|
||||
| [Quickstart](https://yakhyo.github.io/uniface/quickstart/) | Get up and running in 5 minutes |
|
||||
| [Model Zoo](https://yakhyo.github.io/uniface/models/) | All models, benchmarks, and selection guide |
|
||||
| [API Reference](https://yakhyo.github.io/uniface/modules/detection/) | Detailed module documentation |
|
||||
| [Tutorials](https://yakhyo.github.io/uniface/recipes/image-pipeline/) | Step-by-step workflow examples |
|
||||
| [Guides](https://yakhyo.github.io/uniface/concepts/overview/) | Architecture and design principles |
|
||||
| [Datasets](https://yakhyo.github.io/uniface/datasets/) | Training data and evaluation benchmarks |
|
||||
|
||||
---
|
||||
|
||||
## Datasets
|
||||
|
||||
| Task | Training Dataset | Models |
|
||||
|------|-----------------|--------|
|
||||
| Detection | WIDER FACE | RetinaFace, SCRFD, YOLOv5-Face, YOLOv8-Face |
|
||||
| Recognition | MS1MV2 | MobileFace, SphereFace |
|
||||
| Recognition | WebFace600K | ArcFace |
|
||||
| Recognition | WebFace4M / 12M | AdaFace |
|
||||
| Gaze | Gaze360 | MobileGaze |
|
||||
| Head Pose | 300W-LP | HeadPose (ResNet, MobileNet) |
|
||||
| Parsing | CelebAMask-HQ | BiSeNet |
|
||||
| Attributes | CelebA, FairFace, AffectNet | AgeGender, FairFace, Emotion |
|
||||
|
||||
> See [Datasets documentation](https://yakhyo.github.io/uniface/datasets/) for download links, benchmarks, and details.
|
||||
|
||||
---
|
||||
|
||||
## Jupyter Notebooks
|
||||
@@ -209,6 +171,53 @@ Full documentation: https://yakhyo.github.io/uniface/
|
||||
| [09_face_segmentation.ipynb](examples/09_face_segmentation.ipynb) | [](https://colab.research.google.com/github/yakhyo/uniface/blob/main/examples/09_face_segmentation.ipynb) | Face segmentation with XSeg |
|
||||
| [10_face_vector_store.ipynb](examples/10_face_vector_store.ipynb) | [](https://colab.research.google.com/github/yakhyo/uniface/blob/main/examples/10_face_vector_store.ipynb) | FAISS-backed face database |
|
||||
| [11_head_pose_estimation.ipynb](examples/11_head_pose_estimation.ipynb) | [](https://colab.research.google.com/github/yakhyo/uniface/blob/main/examples/11_head_pose_estimation.ipynb) | Head pose estimation (pitch, yaw, roll) |
|
||||
| [12_face_recognition.ipynb](examples/12_face_recognition.ipynb) | [](https://colab.research.google.com/github/yakhyo/uniface/blob/main/examples/12_face_recognition.ipynb) | Standalone face recognition pipeline |
|
||||
|
||||
---
|
||||
|
||||
## Documentation
|
||||
|
||||
Full documentation: https://yakhyo.github.io/uniface/
|
||||
|
||||
| Resource | Description |
|
||||
|----------|-------------|
|
||||
| [Quickstart](https://yakhyo.github.io/uniface/quickstart/) | Get up and running in 5 minutes |
|
||||
| [Model Zoo](https://yakhyo.github.io/uniface/models/) | All models, benchmarks, and selection guide |
|
||||
| [API Reference](https://yakhyo.github.io/uniface/modules/detection/) | Detailed module documentation |
|
||||
| [Tutorials](https://yakhyo.github.io/uniface/recipes/image-pipeline/) | Step-by-step workflow examples |
|
||||
| [Guides](https://yakhyo.github.io/uniface/concepts/overview/) | Architecture and design principles |
|
||||
| [Datasets](https://yakhyo.github.io/uniface/datasets/) | Training data and evaluation benchmarks |
|
||||
|
||||
---
|
||||
|
||||
## Execution Providers (ONNX Runtime)
|
||||
|
||||
```python
|
||||
from uniface.detection import RetinaFace
|
||||
|
||||
# Force CPU-only inference
|
||||
detector = RetinaFace(providers=["CPUExecutionProvider"])
|
||||
```
|
||||
|
||||
See more in the docs:
|
||||
https://yakhyo.github.io/uniface/concepts/execution-providers/
|
||||
|
||||
---
|
||||
|
||||
## Datasets
|
||||
|
||||
| Task | Training Dataset | Models |
|
||||
|------|-----------------|--------|
|
||||
| Detection | WIDER FACE | RetinaFace, SCRFD, YOLOv5-Face, YOLOv8-Face |
|
||||
| Recognition | MS1MV2 | MobileFace, SphereFace |
|
||||
| Recognition | WebFace600K | ArcFace |
|
||||
| Recognition | WebFace4M / 12M | AdaFace |
|
||||
| Gaze | Gaze360 | MobileGaze |
|
||||
| Head Pose | 300W-LP | HeadPose (ResNet, MobileNet) |
|
||||
| Parsing | CelebAMask-HQ | BiSeNet |
|
||||
| Attributes | CelebA, FairFace, AffectNet | AgeGender, FairFace, Emotion |
|
||||
|
||||
> See [Datasets documentation](https://yakhyo.github.io/uniface/datasets/) for download links, benchmarks, and details.
|
||||
|
||||
---
|
||||
|
||||
|
||||
BIN
assets/test_images/image5.jpg
Normal file
BIN
assets/test_images/image5.jpg
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 5.8 KiB |
@@ -33,7 +33,7 @@ graph TB
|
||||
TRK[BYTETracker]
|
||||
end
|
||||
|
||||
subgraph Indexing
|
||||
subgraph Stores
|
||||
IDX[FAISS Vector Store]
|
||||
end
|
||||
|
||||
@@ -124,7 +124,7 @@ uniface/
|
||||
├── headpose/ # Head pose estimation
|
||||
├── spoofing/ # Anti-spoofing
|
||||
├── privacy/ # Face anonymization
|
||||
├── indexing/ # Vector indexing (FAISS)
|
||||
├── stores/ # Vector stores (FAISS)
|
||||
├── types.py # Dataclasses (Face, GazeResult, HeadPoseResult, etc.)
|
||||
├── constants.py # Model weights and URLs
|
||||
├── model_store.py # Model download and caching
|
||||
|
||||
@@ -201,17 +201,11 @@ For drawing detections, filter by confidence:
|
||||
```python
|
||||
from uniface.draw import draw_detections
|
||||
|
||||
# Only draw high-confidence detections
|
||||
bboxes = [f.bbox for f in faces if f.confidence > 0.7]
|
||||
scores = [f.confidence for f in faces if f.confidence > 0.7]
|
||||
landmarks = [f.landmarks for f in faces if f.confidence > 0.7]
|
||||
|
||||
# Only draw high-confidence detections (confidence ≥ vis_threshold)
|
||||
draw_detections(
|
||||
image=image,
|
||||
bboxes=bboxes,
|
||||
scores=scores,
|
||||
landmarks=landmarks,
|
||||
vis_threshold=0.6 # Additional visualization filter
|
||||
faces=faces,
|
||||
vis_threshold=0.7,
|
||||
)
|
||||
```
|
||||
|
||||
|
||||
@@ -32,7 +32,7 @@ ruff check . --fix
|
||||
**Guidelines:**
|
||||
|
||||
- Line length: 120
|
||||
- Python 3.11+ type hints
|
||||
- Python 3.10+ type hints
|
||||
- Google-style docstrings
|
||||
|
||||
---
|
||||
|
||||
@@ -13,7 +13,7 @@ template: home.html
|
||||
<p class="hero-subtitle">All-in-One Open-Source Face Analysis Library</p>
|
||||
|
||||
[](https://pypi.org/project/uniface/)
|
||||
[](https://www.python.org/)
|
||||
[](https://www.python.org/)
|
||||
[](https://opensource.org/licenses/MIT)
|
||||
[](https://github.com/yakhyo/uniface/actions)
|
||||
[](https://pepy.tech/projects/uniface)
|
||||
|
||||
@@ -6,7 +6,7 @@ This guide covers all installation options for UniFace.
|
||||
|
||||
## Requirements
|
||||
|
||||
- **Python**: 3.11 or higher
|
||||
- **Python**: 3.10 or higher
|
||||
- **Operating Systems**: macOS, Linux, Windows
|
||||
|
||||
---
|
||||
@@ -70,16 +70,16 @@ print("Available providers:", ort.get_available_providers())
|
||||
|
||||
---
|
||||
|
||||
### FAISS Vector Indexing
|
||||
### FAISS Vector Store
|
||||
|
||||
For fast multi-identity face search using a FAISS index:
|
||||
For fast multi-identity face search using a FAISS vector store:
|
||||
|
||||
```bash
|
||||
pip install faiss-cpu # CPU
|
||||
pip install faiss-gpu # NVIDIA GPU (CUDA)
|
||||
```
|
||||
|
||||
See the [Indexing module](modules/indexing.md) for usage.
|
||||
See the [Stores module](modules/stores.md) for usage.
|
||||
|
||||
---
|
||||
|
||||
@@ -128,7 +128,7 @@ UniFace has minimal dependencies:
|
||||
|
||||
| Package | Install extra | Purpose |
|
||||
|---------|---------------|---------|
|
||||
| `faiss-cpu` / `faiss-gpu` | `pip install faiss-cpu` | FAISS vector indexing |
|
||||
| `faiss-cpu` / `faiss-gpu` | `pip install faiss-cpu` | FAISS vector store |
|
||||
| `onnxruntime-gpu` | `uniface[gpu]` | CUDA acceleration |
|
||||
| `torch` | `pip install torch` | Emotion model uses TorchScript |
|
||||
| `torchvision` | `pip install torchvision` | Faster NMS for YOLO detectors |
|
||||
@@ -159,11 +159,11 @@ print("Installation successful!")
|
||||
|
||||
### Import Errors
|
||||
|
||||
If you encounter import errors, ensure you're using Python 3.11+:
|
||||
If you encounter import errors, ensure you're using Python 3.10+:
|
||||
|
||||
```bash
|
||||
python --version
|
||||
# Should show: Python 3.11.x or higher
|
||||
# Should show: Python 3.10.x or higher
|
||||
```
|
||||
|
||||
### Model Download Issues
|
||||
|
||||
@@ -264,10 +264,8 @@ from uniface.draw import draw_detections
|
||||
|
||||
draw_detections(
|
||||
image=image,
|
||||
bboxes=[f.bbox for f in faces],
|
||||
scores=[f.confidence for f in faces],
|
||||
landmarks=[f.landmarks for f in faces],
|
||||
vis_threshold=0.6
|
||||
faces=faces,
|
||||
vis_threshold=0.6,
|
||||
)
|
||||
|
||||
cv2.imwrite("result.jpg", image)
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# Indexing
|
||||
# Stores
|
||||
|
||||
FAISS-backed vector store for fast similarity search over embeddings.
|
||||
|
||||
@@ -12,7 +12,7 @@ FAISS-backed vector store for fast similarity search over embeddings.
|
||||
## FAISS
|
||||
|
||||
```python
|
||||
from uniface.indexing import FAISS
|
||||
from uniface.stores import FAISS
|
||||
```
|
||||
|
||||
A thin wrapper around a FAISS `IndexFlatIP` (inner-product) index. Vectors
|
||||
@@ -134,7 +134,7 @@ loaded = store.load() # True if files exist
|
||||
import cv2
|
||||
from uniface.detection import RetinaFace
|
||||
from uniface.recognition import ArcFace
|
||||
from uniface.indexing import FAISS
|
||||
from uniface.stores import FAISS
|
||||
|
||||
detector = RetinaFace()
|
||||
recognizer = ArcFace()
|
||||
@@ -19,6 +19,7 @@ Run UniFace examples directly in your browser with Google Colab, or download and
|
||||
| [Face Segmentation](https://github.com/yakhyo/uniface/blob/main/examples/09_face_segmentation.ipynb) | [](https://colab.research.google.com/github/yakhyo/uniface/blob/main/examples/09_face_segmentation.ipynb) | Face segmentation with XSeg |
|
||||
| [Face Vector Store](https://github.com/yakhyo/uniface/blob/main/examples/10_face_vector_store.ipynb) | [](https://colab.research.google.com/github/yakhyo/uniface/blob/main/examples/10_face_vector_store.ipynb) | FAISS-backed face database |
|
||||
| [Head Pose Estimation](https://github.com/yakhyo/uniface/blob/main/examples/11_head_pose_estimation.ipynb) | [](https://colab.research.google.com/github/yakhyo/uniface/blob/main/examples/11_head_pose_estimation.ipynb) | 3D head orientation estimation |
|
||||
| [Face Recognition](https://github.com/yakhyo/uniface/blob/main/examples/12_face_recognition.ipynb) | [](https://colab.research.google.com/github/yakhyo/uniface/blob/main/examples/12_face_recognition.ipynb) | Standalone face recognition pipeline |
|
||||
|
||||
---
|
||||
|
||||
|
||||
@@ -54,19 +54,8 @@ detector = RetinaFace()
|
||||
image = cv2.imread("photo.jpg")
|
||||
faces = detector.detect(image)
|
||||
|
||||
# Extract visualization data
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
|
||||
# Draw on image
|
||||
draw_detections(
|
||||
image=image,
|
||||
bboxes=bboxes,
|
||||
scores=scores,
|
||||
landmarks=landmarks,
|
||||
vis_threshold=0.6,
|
||||
)
|
||||
draw_detections(image=image, faces=faces, vis_threshold=0.6)
|
||||
|
||||
# Save result
|
||||
cv2.imwrite("output.jpg", image)
|
||||
@@ -372,10 +361,7 @@ while True:
|
||||
|
||||
faces = detector.detect(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks)
|
||||
draw_detections(image=frame, faces=faces)
|
||||
|
||||
cv2.imshow("UniFace - Press 'q' to quit", frame)
|
||||
|
||||
@@ -507,7 +493,7 @@ from uniface.privacy import BlurFace
|
||||
from uniface.spoofing import MiniFASNet
|
||||
from uniface.tracking import BYTETracker
|
||||
from uniface.analyzer import FaceAnalyzer
|
||||
from uniface.indexing import FAISS # pip install faiss-cpu
|
||||
from uniface.stores import FAISS # pip install faiss-cpu
|
||||
from uniface.draw import draw_detections, draw_tracks
|
||||
```
|
||||
|
||||
|
||||
@@ -52,7 +52,7 @@ python tools/search.py --reference ref.jpg --source 0 # webcam
|
||||
## Vector Search (FAISS index)
|
||||
|
||||
For identifying faces against a database of many known people, use the
|
||||
[`FAISS`](../modules/indexing.md) vector store.
|
||||
[`FAISS`](../modules/stores.md) vector store.
|
||||
|
||||
!!! info "Install extra"
|
||||
`bash
|
||||
@@ -80,7 +80,7 @@ import cv2
|
||||
from pathlib import Path
|
||||
from uniface.detection import RetinaFace
|
||||
from uniface.recognition import ArcFace
|
||||
from uniface.indexing import FAISS
|
||||
from uniface.stores import FAISS
|
||||
|
||||
detector = RetinaFace()
|
||||
recognizer = ArcFace()
|
||||
@@ -112,7 +112,7 @@ python tools/faiss_search.py build --faces-dir dataset/ --db-path ./my_index
|
||||
import cv2
|
||||
from uniface.detection import RetinaFace
|
||||
from uniface.recognition import ArcFace
|
||||
from uniface.indexing import FAISS
|
||||
from uniface.stores import FAISS
|
||||
|
||||
detector = RetinaFace()
|
||||
recognizer = ArcFace()
|
||||
@@ -143,7 +143,7 @@ python tools/faiss_search.py run --db-path ./my_index --source 0 # webcam
|
||||
### Manage the index
|
||||
|
||||
```python
|
||||
from uniface.indexing import FAISS
|
||||
from uniface.stores import FAISS
|
||||
|
||||
store = FAISS(db_path="./my_index")
|
||||
store.load()
|
||||
@@ -160,7 +160,7 @@ store.save()
|
||||
|
||||
## See Also
|
||||
|
||||
- [Indexing Module](../modules/indexing.md) - Full `FAISS` API reference
|
||||
- [Stores Module](../modules/stores.md) - Full `FAISS` API reference
|
||||
- [Recognition Module](../modules/recognition.md) - Face recognition details
|
||||
- [Video & Webcam](video-webcam.md) - Real-time processing
|
||||
- [Concepts: Thresholds](../concepts/thresholds-calibration.md) - Tuning similarity thresholds
|
||||
|
||||
@@ -48,12 +48,7 @@ def process_image(image_path):
|
||||
print(f" Face {i+1}: {attrs.sex}, {attrs.age} years old")
|
||||
|
||||
# Visualize
|
||||
draw_detections(
|
||||
image=image,
|
||||
bboxes=[f.bbox for f in faces],
|
||||
scores=[f.confidence for f in faces],
|
||||
landmarks=[f.landmarks for f in faces]
|
||||
)
|
||||
draw_detections(image=image, faces=faces)
|
||||
|
||||
return image, results
|
||||
|
||||
|
||||
@@ -26,12 +26,7 @@ while True:
|
||||
|
||||
faces = detector.detect(frame)
|
||||
|
||||
draw_detections(
|
||||
image=frame,
|
||||
bboxes=[f.bbox for f in faces],
|
||||
scores=[f.confidence for f in faces],
|
||||
landmarks=[f.landmarks for f in faces]
|
||||
)
|
||||
draw_detections(image=frame, faces=faces)
|
||||
|
||||
cv2.imshow("Face Detection", frame)
|
||||
|
||||
|
||||
@@ -51,7 +51,7 @@
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"3.2.0\n"
|
||||
"3.3.0\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
@@ -156,13 +156,8 @@
|
||||
"faces = detector.detect(image)\n",
|
||||
"print(f'Detected {len(faces)} face(s)')\n",
|
||||
"\n",
|
||||
"# Unpack face data for visualization\n",
|
||||
"bboxes = [f.bbox for f in faces]\n",
|
||||
"scores = [f.confidence for f in faces]\n",
|
||||
"landmarks = [f.landmarks for f in faces]\n",
|
||||
"\n",
|
||||
"# Draw detections\n",
|
||||
"draw_detections(image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=0.6, corner_bbox=True)\n",
|
||||
"draw_detections(image=image, faces=faces, vis_threshold=0.6, corner_bbox=True)\n",
|
||||
"\n",
|
||||
"# Display result\n",
|
||||
"output_image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
|
||||
@@ -210,11 +205,7 @@
|
||||
"faces = detector.detect(image, max_num=2)\n",
|
||||
"print(f'Detected {len(faces)} face(s)')\n",
|
||||
"\n",
|
||||
"bboxes = [f.bbox for f in faces]\n",
|
||||
"scores = [f.confidence for f in faces]\n",
|
||||
"landmarks = [f.landmarks for f in faces]\n",
|
||||
"\n",
|
||||
"draw_detections(image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=0.6, corner_bbox=True)\n",
|
||||
"draw_detections(image=image, faces=faces, vis_threshold=0.6, corner_bbox=True)\n",
|
||||
"\n",
|
||||
"output_image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
|
||||
"display.display(Image.fromarray(output_image))"
|
||||
@@ -257,11 +248,7 @@
|
||||
"faces = detector.detect(image, max_num=5)\n",
|
||||
"print(f'Detected {len(faces)} face(s)')\n",
|
||||
"\n",
|
||||
"bboxes = [f.bbox for f in faces]\n",
|
||||
"scores = [f.confidence for f in faces]\n",
|
||||
"landmarks = [f.landmarks for f in faces]\n",
|
||||
"\n",
|
||||
"draw_detections(image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=0.6, corner_bbox=True)\n",
|
||||
"draw_detections(image=image, faces=faces, vis_threshold=0.6, corner_bbox=True)\n",
|
||||
"\n",
|
||||
"output_image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
|
||||
"display.display(Image.fromarray(output_image))"
|
||||
|
||||
@@ -55,7 +55,7 @@
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"3.2.0\n"
|
||||
"3.3.0\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
@@ -139,10 +139,7 @@
|
||||
"\n",
|
||||
" # Draw detections\n",
|
||||
" bbox_image = image.copy()\n",
|
||||
" bboxes = [f.bbox for f in faces]\n",
|
||||
" scores = [f.confidence for f in faces]\n",
|
||||
" landmarks = [f.landmarks for f in faces]\n",
|
||||
" draw_detections(image=bbox_image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=0.6, corner_bbox=True)\n",
|
||||
" draw_detections(image=bbox_image, faces=faces, vis_threshold=0.6, corner_bbox=True)\n",
|
||||
"\n",
|
||||
" # Align first detected face (returns aligned image and inverse transform matrix)\n",
|
||||
" first_landmarks = faces[0].landmarks\n",
|
||||
|
||||
@@ -44,7 +44,7 @@
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"3.2.0\n"
|
||||
"3.3.0\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
|
||||
@@ -49,7 +49,7 @@
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"3.2.0\n"
|
||||
"3.3.0\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
|
||||
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
@@ -51,7 +51,7 @@
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"UniFace version: 3.2.0\n"
|
||||
"UniFace version: 3.3.0\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
|
||||
File diff suppressed because one or more lines are too long
@@ -53,7 +53,7 @@
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"UniFace version: 3.2.0\n"
|
||||
"UniFace version: 3.3.0\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
|
||||
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
422
examples/12_face_recognition.ipynb
Normal file
422
examples/12_face_recognition.ipynb
Normal file
File diff suppressed because one or more lines are too long
@@ -154,7 +154,7 @@ nav:
|
||||
- Head Pose: modules/headpose.md
|
||||
- Anti-Spoofing: modules/spoofing.md
|
||||
- Privacy: modules/privacy.md
|
||||
- Indexing: modules/indexing.md
|
||||
- Stores: modules/stores.md
|
||||
- Guides:
|
||||
- Overview: concepts/overview.md
|
||||
- Inputs & Outputs: concepts/inputs-outputs.md
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
[project]
|
||||
name = "uniface"
|
||||
version = "3.2.0"
|
||||
version = "3.3.0"
|
||||
description = "UniFace: A Comprehensive Library for Face Detection, Recognition, Tracking, Landmark Analysis, Face Parsing, Gaze Estimation, Age, and Gender Detection"
|
||||
readme = "README.md"
|
||||
license = "MIT"
|
||||
@@ -9,7 +9,7 @@ maintainers = [
|
||||
{ name = "Yakhyokhuja Valikhujaev", email = "yakhyo9696@gmail.com" },
|
||||
]
|
||||
|
||||
requires-python = ">=3.11,<3.15"
|
||||
requires-python = ">=3.10,<3.15"
|
||||
keywords = [
|
||||
"face-detection",
|
||||
"face-recognition",
|
||||
@@ -34,6 +34,7 @@ classifiers = [
|
||||
"Intended Audience :: Science/Research",
|
||||
"Operating System :: OS Independent",
|
||||
"Programming Language :: Python :: 3",
|
||||
"Programming Language :: Python :: 3.10",
|
||||
"Programming Language :: Python :: 3.11",
|
||||
"Programming Language :: Python :: 3.12",
|
||||
"Programming Language :: Python :: 3.13",
|
||||
@@ -44,7 +45,7 @@ dependencies = [
|
||||
"numpy>=1.21.0",
|
||||
"opencv-python>=4.5.0",
|
||||
"onnxruntime>=1.16.0",
|
||||
"scikit-image>=0.26.0",
|
||||
"scikit-image>=0.22.0",
|
||||
"scipy>=1.7.0",
|
||||
"requests>=2.28.0",
|
||||
"tqdm>=4.64.0",
|
||||
@@ -73,7 +74,7 @@ uniface = ["py.typed"]
|
||||
|
||||
[tool.ruff]
|
||||
line-length = 120
|
||||
target-version = "py311"
|
||||
target-version = "py310"
|
||||
exclude = [
|
||||
".git",
|
||||
".ruff_cache",
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
numpy>=1.21.0
|
||||
opencv-python>=4.5.0
|
||||
onnxruntime>=1.16.0
|
||||
scikit-image>=0.26.0
|
||||
scikit-image>=0.22.0
|
||||
scipy>=1.7.0
|
||||
requests>=2.28.0
|
||||
tqdm>=4.64.0
|
||||
|
||||
@@ -27,12 +27,17 @@ from uniface.draw import draw_detections
|
||||
from uniface.recognition import ArcFace
|
||||
|
||||
|
||||
def draw_face_info(image, face, face_id):
|
||||
"""Draw face ID and attributes above bounding box."""
|
||||
def draw_face_info(image, face):
|
||||
"""Draw face attributes above bounding box."""
|
||||
x1, y1, _x2, y2 = map(int, face.bbox)
|
||||
lines = [f'ID: {face_id}', f'Conf: {face.confidence:.2f}']
|
||||
if face.age and face.sex:
|
||||
lines = []
|
||||
if face.age is not None and face.sex is not None:
|
||||
lines.append(f'{face.sex}, {face.age}y')
|
||||
if face.emotion is not None:
|
||||
lines.append(face.emotion)
|
||||
|
||||
if not lines:
|
||||
return
|
||||
|
||||
for i, line in enumerate(lines):
|
||||
y_pos = y1 - 10 - (len(lines) - 1 - i) * 25
|
||||
@@ -95,13 +100,10 @@ def process_image(analyzer, image_path: str, save_dir: str = 'outputs', show_sim
|
||||
status = 'Same' if sim > 0.4 else 'Different'
|
||||
print(f' Face {i + 1} ↔ Face {j + 1}: {sim:.3f} ({status})')
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, corner_bbox=True)
|
||||
draw_detections(image=image, faces=faces, corner_bbox=True)
|
||||
|
||||
for i, face in enumerate(faces, 1):
|
||||
draw_face_info(image, face, i)
|
||||
for face in faces:
|
||||
draw_face_info(image, face)
|
||||
|
||||
os.makedirs(save_dir, exist_ok=True)
|
||||
output_path = os.path.join(save_dir, f'{Path(image_path).stem}_analysis.jpg')
|
||||
@@ -137,13 +139,10 @@ def process_video(analyzer, video_path: str, save_dir: str = 'outputs'):
|
||||
frame_count += 1
|
||||
faces = analyzer.analyze(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks, corner_bbox=True)
|
||||
draw_detections(image=frame, faces=faces, corner_bbox=True)
|
||||
|
||||
for i, face in enumerate(faces, 1):
|
||||
draw_face_info(frame, face, i)
|
||||
for face in faces:
|
||||
draw_face_info(frame, face)
|
||||
|
||||
cv2.putText(frame, f'Faces: {len(faces)}', (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)
|
||||
out.write(frame)
|
||||
@@ -173,13 +172,10 @@ def run_camera(analyzer, camera_id: int = 0):
|
||||
|
||||
faces = analyzer.analyze(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks, corner_bbox=True)
|
||||
draw_detections(image=frame, faces=faces, corner_bbox=True)
|
||||
|
||||
for i, face in enumerate(faces, 1):
|
||||
draw_face_info(frame, face, i)
|
||||
for face in faces:
|
||||
draw_face_info(frame, face)
|
||||
|
||||
cv2.putText(frame, f'Faces: {len(faces)}', (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)
|
||||
cv2.imshow('Face Analyzer', frame)
|
||||
|
||||
@@ -43,10 +43,7 @@ def process_image(
|
||||
from uniface.draw import draw_detections
|
||||
|
||||
preview = image.copy()
|
||||
bboxes = [face.bbox for face in faces]
|
||||
scores = [face.confidence for face in faces]
|
||||
landmarks = [face.landmarks for face in faces]
|
||||
draw_detections(preview, bboxes, scores, landmarks)
|
||||
draw_detections(image=preview, faces=faces)
|
||||
|
||||
cv2.imshow('Detections (Press any key to continue)', preview)
|
||||
cv2.waitKey(0)
|
||||
|
||||
@@ -52,12 +52,7 @@ def process_image(
|
||||
if not faces:
|
||||
return
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=image, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for i, face in enumerate(faces):
|
||||
result = age_gender.predict(image, face)
|
||||
@@ -104,12 +99,7 @@ def process_video(
|
||||
frame_count += 1
|
||||
faces = detector.detect(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=frame, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for face in faces:
|
||||
result = age_gender.predict(frame, face)
|
||||
@@ -143,12 +133,7 @@ def run_camera(detector, age_gender, camera_id: int = 0, threshold: float = 0.6)
|
||||
|
||||
faces = detector.detect(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=frame, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for face in faces:
|
||||
result = age_gender.predict(frame, face)
|
||||
|
||||
@@ -34,13 +34,7 @@ def process_image(detector, image_path: Path, output_path: Path, threshold: floa
|
||||
|
||||
faces = detector.detect(image)
|
||||
|
||||
# unpack face data for visualization
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=image, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
cv2.putText(
|
||||
image,
|
||||
|
||||
@@ -35,10 +35,7 @@ def process_image(detector, image_path: str, threshold: float = 0.6, save_dir: s
|
||||
faces = detector.detect(image)
|
||||
|
||||
if faces:
|
||||
bboxes = [face.bbox for face in faces]
|
||||
scores = [face.confidence for face in faces]
|
||||
landmarks = [face.landmarks for face in faces]
|
||||
draw_detections(image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold)
|
||||
draw_detections(image=image, faces=faces, vis_threshold=threshold)
|
||||
|
||||
os.makedirs(save_dir, exist_ok=True)
|
||||
output_path = os.path.join(save_dir, f'{os.path.splitext(os.path.basename(image_path))[0]}_out.jpg')
|
||||
@@ -89,14 +86,9 @@ def process_video(
|
||||
faces = detector.detect(frame)
|
||||
total_faces += len(faces)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=frame,
|
||||
bboxes=bboxes,
|
||||
scores=scores,
|
||||
landmarks=landmarks,
|
||||
faces=faces,
|
||||
vis_threshold=threshold,
|
||||
draw_score=True,
|
||||
corner_bbox=True,
|
||||
@@ -141,14 +133,9 @@ def run_camera(detector, camera_id: int = 0, threshold: float = 0.6):
|
||||
|
||||
faces = detector.detect(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=frame,
|
||||
bboxes=bboxes,
|
||||
scores=scores,
|
||||
landmarks=landmarks,
|
||||
faces=faces,
|
||||
vis_threshold=threshold,
|
||||
draw_score=True,
|
||||
corner_bbox=True,
|
||||
|
||||
@@ -52,12 +52,7 @@ def process_image(
|
||||
if not faces:
|
||||
return
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=image, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for i, face in enumerate(faces):
|
||||
result = emotion_predictor.predict(image, face)
|
||||
@@ -104,12 +99,7 @@ def process_video(
|
||||
frame_count += 1
|
||||
faces = detector.detect(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=frame, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for face in faces:
|
||||
result = emotion_predictor.predict(frame, face)
|
||||
@@ -143,12 +133,7 @@ def run_camera(detector, emotion_predictor, camera_id: int = 0, threshold: float
|
||||
|
||||
faces = detector.detect(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=frame, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for face in faces:
|
||||
result = emotion_predictor.predict(frame, face)
|
||||
|
||||
@@ -52,12 +52,7 @@ def process_image(
|
||||
if not faces:
|
||||
return
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=image, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=image, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for i, face in enumerate(faces):
|
||||
result = fairface.predict(image, face)
|
||||
@@ -104,12 +99,7 @@ def process_video(
|
||||
frame_count += 1
|
||||
faces = detector.detect(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=frame, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for face in faces:
|
||||
result = fairface.predict(frame, face)
|
||||
@@ -143,12 +133,7 @@ def run_camera(detector, fairface, camera_id: int = 0, threshold: float = 0.6):
|
||||
|
||||
faces = detector.detect(frame)
|
||||
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
draw_detections(
|
||||
image=frame, bboxes=bboxes, scores=scores, landmarks=landmarks, vis_threshold=threshold, corner_bbox=True
|
||||
)
|
||||
draw_detections(image=frame, faces=faces, vis_threshold=threshold, corner_bbox=True)
|
||||
|
||||
for face in faces:
|
||||
result = fairface.predict(frame, face)
|
||||
|
||||
@@ -24,7 +24,7 @@ import cv2
|
||||
|
||||
from uniface import create_detector, create_recognizer
|
||||
from uniface.draw import draw_corner_bbox, draw_text_label
|
||||
from uniface.indexing import FAISS
|
||||
from uniface.stores import FAISS
|
||||
|
||||
|
||||
def _draw_face(image, bbox, text: str, color: tuple[int, int, int]) -> None:
|
||||
|
||||
@@ -30,7 +30,7 @@ from __future__ import annotations
|
||||
|
||||
__license__ = 'MIT'
|
||||
__author__ = 'Yakhyokhuja Valikhujaev'
|
||||
__version__ = '3.2.0'
|
||||
__version__ = '3.3.0'
|
||||
|
||||
import contextlib
|
||||
|
||||
@@ -60,7 +60,7 @@ from .types import AttributeResult, EmotionResult, Face, GazeResult, HeadPoseRes
|
||||
|
||||
# Optional: FAISS vector store (requires `pip install faiss-cpu`)
|
||||
with contextlib.suppress(ImportError):
|
||||
from .indexing import FAISS
|
||||
from .stores import FAISS
|
||||
|
||||
__all__ = [
|
||||
# Metadata
|
||||
@@ -114,7 +114,7 @@ __all__ = [
|
||||
'BYTETracker',
|
||||
# Privacy
|
||||
'BlurFace',
|
||||
# Indexing (optional)
|
||||
# Stores (optional)
|
||||
'FAISS',
|
||||
# Utilities
|
||||
'Logger',
|
||||
|
||||
@@ -4,6 +4,8 @@
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
from typing import Any
|
||||
|
||||
import numpy as np
|
||||
|
||||
from uniface.attribute.base import Attribute
|
||||
@@ -14,6 +16,8 @@ from uniface.types import Face
|
||||
|
||||
__all__ = ['FaceAnalyzer']
|
||||
|
||||
_UNSET: Any = object()
|
||||
|
||||
|
||||
class FaceAnalyzer:
|
||||
"""Unified face analyzer combining detection, recognition, and attributes.
|
||||
@@ -27,35 +31,52 @@ class FaceAnalyzer:
|
||||
via the ``attributes`` list. Each predictor's ``predict(image, face)``
|
||||
is called once per detected face, enriching the :class:`Face` in-place.
|
||||
|
||||
Args:
|
||||
detector: Face detector instance for detecting faces in images.
|
||||
recognizer: Optional face recognizer for extracting embeddings.
|
||||
attributes: Optional list of ``Attribute`` predictors to run on
|
||||
each detected face (e.g. ``[AgeGender(), FairFace(), Emotion()]``).
|
||||
When called with no arguments, uses SCRFD (500M) for detection and
|
||||
ArcFace (MobileNet) for recognition — the smallest and fastest variants.
|
||||
|
||||
Example:
|
||||
>>> from uniface import RetinaFace, ArcFace, AgeGender, FaceAnalyzer
|
||||
>>> detector = RetinaFace()
|
||||
>>> recognizer = ArcFace()
|
||||
>>> analyzer = FaceAnalyzer(detector, recognizer=recognizer, attributes=[AgeGender()])
|
||||
Args:
|
||||
detector: Face detector instance. Defaults to ``SCRFD(SCRFD_500M_KPS)``.
|
||||
recognizer: Face recognizer for extracting embeddings.
|
||||
Defaults to ``ArcFace(MNET)``. Pass ``None`` to disable recognition.
|
||||
attributes: Optional list of ``Attribute`` predictors to run on
|
||||
each detected face (e.g. ``[AgeGender()]``).
|
||||
|
||||
Examples:
|
||||
>>> from uniface import FaceAnalyzer
|
||||
>>> analyzer = FaceAnalyzer()
|
||||
>>> faces = analyzer.analyze(image)
|
||||
|
||||
>>> from uniface import FaceAnalyzer, AgeGender
|
||||
>>> analyzer = FaceAnalyzer(attributes=[AgeGender()])
|
||||
>>> faces = analyzer.analyze(image)
|
||||
"""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
detector: BaseDetector,
|
||||
recognizer: BaseRecognizer | None = None,
|
||||
detector: BaseDetector | None = None,
|
||||
recognizer: BaseRecognizer | None = _UNSET,
|
||||
attributes: list[Attribute] | None = None,
|
||||
) -> None:
|
||||
if detector is None:
|
||||
from uniface.constants import SCRFDWeights
|
||||
from uniface.detection import SCRFD
|
||||
|
||||
detector = SCRFD(model_name=SCRFDWeights.SCRFD_500M_KPS)
|
||||
|
||||
if recognizer is _UNSET:
|
||||
from uniface.recognition import ArcFace
|
||||
|
||||
recognizer = ArcFace()
|
||||
|
||||
self.detector = detector
|
||||
self.recognizer = recognizer
|
||||
self.attributes: list[Attribute] = attributes or []
|
||||
|
||||
Logger.info(f'Initialized FaceAnalyzer with detector={detector.__class__.__name__}')
|
||||
if recognizer:
|
||||
Logger.info(f' - Recognition enabled: {recognizer.__class__.__name__}')
|
||||
Logger.info(f'Recognition enabled: {recognizer.__class__.__name__}')
|
||||
for attr in self.attributes:
|
||||
Logger.info(f' - Attribute enabled: {attr.__class__.__name__}')
|
||||
Logger.info(f'Attribute enabled: {attr.__class__.__name__}')
|
||||
|
||||
def analyze(self, image: np.ndarray) -> list[Face]:
|
||||
"""Analyze faces in an image.
|
||||
@@ -76,17 +97,17 @@ class FaceAnalyzer:
|
||||
if self.recognizer is not None:
|
||||
try:
|
||||
face.embedding = self.recognizer.get_normalized_embedding(image, face.landmarks)
|
||||
Logger.debug(f' Face {idx + 1}: Extracted embedding with shape {face.embedding.shape}')
|
||||
Logger.debug(f'Face {idx + 1}: Extracted embedding with shape {face.embedding.shape}')
|
||||
except Exception as e:
|
||||
Logger.warning(f' Face {idx + 1}: Failed to extract embedding: {e}')
|
||||
Logger.warning(f'Face {idx + 1}: Failed to extract embedding: {e}')
|
||||
|
||||
for attr in self.attributes:
|
||||
attr_name = attr.__class__.__name__
|
||||
try:
|
||||
attr.predict(image, face)
|
||||
Logger.debug(f' Face {idx + 1}: {attr_name} prediction succeeded')
|
||||
Logger.debug(f'Face {idx + 1}: {attr_name} prediction succeeded')
|
||||
except Exception as e:
|
||||
Logger.warning(f' Face {idx + 1}: {attr_name} prediction failed: {e}')
|
||||
Logger.warning(f'Face {idx + 1}: {attr_name} prediction failed: {e}')
|
||||
|
||||
Logger.info(f'Analysis complete: {len(faces)} face(s) processed')
|
||||
return faces
|
||||
|
||||
@@ -232,9 +232,10 @@ def draw_text_label(
|
||||
def draw_detections(
|
||||
*,
|
||||
image: np.ndarray,
|
||||
bboxes: list[np.ndarray] | list[list[float]],
|
||||
scores: np.ndarray | list[float],
|
||||
landmarks: list[np.ndarray] | list[list[list[float]]],
|
||||
faces: list[Face] | None = None,
|
||||
bboxes: list[np.ndarray] | list[list[float]] | None = None,
|
||||
scores: np.ndarray | list[float] | None = None,
|
||||
landmarks: list[np.ndarray] | list[list[list[float]]] | None = None,
|
||||
vis_threshold: float = 0.6,
|
||||
draw_score: bool = False,
|
||||
corner_bbox: bool = True,
|
||||
@@ -243,17 +244,31 @@ def draw_detections(
|
||||
|
||||
Modifies the image in-place.
|
||||
|
||||
Accepts either a list of :class:`Face` objects (preferred) or separate
|
||||
lists of bboxes, scores, and landmarks for backward compatibility.
|
||||
|
||||
Args:
|
||||
image: Input image to draw on (modified in-place).
|
||||
faces: List of Face objects from detection. When provided,
|
||||
``bboxes``, ``scores``, and ``landmarks`` are ignored.
|
||||
bboxes: List of bounding boxes in xyxy format ``[x1, y1, x2, y2]``.
|
||||
scores: List of confidence scores.
|
||||
landmarks: List of landmark sets with shape ``(5, 2)``.
|
||||
vis_threshold: Confidence threshold for filtering. Defaults to 0.6.
|
||||
draw_score: Whether to draw confidence scores. Defaults to False.
|
||||
corner_bbox: Use corner-style bounding boxes. Defaults to True.
|
||||
"""
|
||||
|
||||
# Adaptive line thickness
|
||||
Examples:
|
||||
>>> draw_detections(image=image, faces=faces)
|
||||
>>> draw_detections(image=image, faces=faces, vis_threshold=0.7, draw_score=True)
|
||||
"""
|
||||
if faces is not None:
|
||||
bboxes = [f.bbox for f in faces]
|
||||
scores = [f.confidence for f in faces]
|
||||
landmarks = [f.landmarks for f in faces]
|
||||
elif bboxes is None or scores is None or landmarks is None:
|
||||
raise ValueError('Provide either faces or all of bboxes, scores, and landmarks')
|
||||
|
||||
line_thickness = max(round(sum(image.shape[:2]) / 2 * 0.003), 2)
|
||||
|
||||
for i, score in enumerate(scores):
|
||||
@@ -262,13 +277,11 @@ def draw_detections(
|
||||
|
||||
bbox = np.array(bboxes[i], dtype=np.int32)
|
||||
|
||||
# Draw bounding box
|
||||
if corner_bbox:
|
||||
draw_corner_bbox(image, bbox, color=(0, 255, 0), thickness=line_thickness, proportion=0.2)
|
||||
else:
|
||||
cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:]), (0, 255, 0), line_thickness)
|
||||
|
||||
# Draw confidence score label
|
||||
if draw_score:
|
||||
font_scale = max(0.4, min(0.7, (bbox[3] - bbox[1]) / 200))
|
||||
draw_text_label(
|
||||
@@ -281,7 +294,6 @@ def draw_detections(
|
||||
font_scale=font_scale,
|
||||
)
|
||||
|
||||
# Draw landmarks
|
||||
landmark_set = np.array(landmarks[i], dtype=np.int32)
|
||||
for j, point in enumerate(landmark_set):
|
||||
cv2.circle(image, tuple(point), line_thickness + 1, _LANDMARK_COLORS[j % len(_LANDMARK_COLORS)], -1)
|
||||
|
||||
@@ -71,7 +71,13 @@ def estimate_norm(
|
||||
alignment[:, 0] += diff_x
|
||||
|
||||
# Compute the transformation matrix
|
||||
transform = SimilarityTransform.from_estimate(landmark, alignment)
|
||||
try:
|
||||
# scikit-image >= 0.26
|
||||
transform = SimilarityTransform.from_estimate(landmark, alignment)
|
||||
except AttributeError:
|
||||
# scikit-image < 0.26 (e.g. Python 3.10 with older scikit-image)
|
||||
transform = SimilarityTransform()
|
||||
transform.estimate(landmark, alignment)
|
||||
|
||||
matrix = transform.params[0:2, :]
|
||||
inverse_matrix = np.linalg.inv(transform.params)[0:2, :]
|
||||
|
||||
@@ -1,9 +0,0 @@
|
||||
# Copyright 2025-2026 Yakhyokhuja Valikhujaev
|
||||
# Author: Yakhyokhuja Valikhujaev
|
||||
# GitHub: https://github.com/yakhyo
|
||||
|
||||
"""Vector indexing backends for fast similarity search."""
|
||||
|
||||
from uniface.indexing.faiss import FAISS
|
||||
|
||||
__all__ = ['FAISS']
|
||||
10
uniface/stores/__init__.py
Normal file
10
uniface/stores/__init__.py
Normal file
@@ -0,0 +1,10 @@
|
||||
# Copyright 2025-2026 Yakhyokhuja Valikhujaev
|
||||
# Author: Yakhyokhuja Valikhujaev
|
||||
# GitHub: https://github.com/yakhyo
|
||||
|
||||
"""Vector store backends for fast face embedding similarity search."""
|
||||
|
||||
from uniface.stores.base import BaseStore
|
||||
from uniface.stores.faiss import FAISS
|
||||
|
||||
__all__ = ['BaseStore', 'FAISS']
|
||||
69
uniface/stores/base.py
Normal file
69
uniface/stores/base.py
Normal file
@@ -0,0 +1,69 @@
|
||||
# Copyright 2025-2026 Yakhyokhuja Valikhujaev
|
||||
# Author: Yakhyokhuja Valikhujaev
|
||||
# GitHub: https://github.com/yakhyo
|
||||
|
||||
"""Abstract base class for vector store backends."""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
from abc import ABC, abstractmethod
|
||||
from typing import Any
|
||||
|
||||
import numpy as np
|
||||
|
||||
__all__ = ['BaseStore']
|
||||
|
||||
Metadata = dict[str, Any]
|
||||
|
||||
|
||||
class BaseStore(ABC):
|
||||
"""Abstract interface for face embedding vector stores.
|
||||
|
||||
All vector store backends (FAISS, Qdrant, etc.) must implement
|
||||
this interface to ensure consistent usage across the library.
|
||||
|
||||
Embeddings are expected to be L2-normalised so that inner product
|
||||
equals cosine similarity.
|
||||
"""
|
||||
|
||||
@abstractmethod
|
||||
def add(self, embedding: np.ndarray, metadata: Metadata) -> None:
|
||||
"""Add a single embedding with associated metadata.
|
||||
|
||||
Args:
|
||||
embedding: L2-normalised embedding vector.
|
||||
metadata: Arbitrary dict of JSON-serialisable key-value pairs.
|
||||
"""
|
||||
|
||||
@abstractmethod
|
||||
def search(
|
||||
self,
|
||||
embedding: np.ndarray,
|
||||
threshold: float = 0.4,
|
||||
) -> tuple[Metadata | None, float]:
|
||||
"""Find the closest match for a query embedding.
|
||||
|
||||
Args:
|
||||
embedding: L2-normalised query vector.
|
||||
threshold: Minimum similarity to accept a match.
|
||||
|
||||
Returns:
|
||||
``(metadata, similarity)`` for the best match, or
|
||||
``(None, similarity)`` when below *threshold* or empty.
|
||||
"""
|
||||
|
||||
@abstractmethod
|
||||
def remove(self, key: str, value: Any) -> int:
|
||||
"""Remove all entries where ``metadata[key] == value``.
|
||||
|
||||
Args:
|
||||
key: Metadata key to match against.
|
||||
value: Value to match.
|
||||
|
||||
Returns:
|
||||
Number of entries removed.
|
||||
"""
|
||||
|
||||
@abstractmethod
|
||||
def __len__(self) -> int:
|
||||
"""Return the number of vectors in the store."""
|
||||
@@ -12,9 +12,9 @@ import numpy as np
|
||||
|
||||
from uniface.log import Logger
|
||||
|
||||
__all__ = ['FAISS']
|
||||
from .base import BaseStore, Metadata
|
||||
|
||||
Metadata = dict[str, Any]
|
||||
__all__ = ['FAISS']
|
||||
|
||||
|
||||
def _import_faiss():
|
||||
@@ -34,7 +34,7 @@ def _import_faiss():
|
||||
return faiss
|
||||
|
||||
|
||||
class FAISS:
|
||||
class FAISS(BaseStore):
|
||||
"""FAISS vector store using IndexFlatIP (inner product).
|
||||
|
||||
Vectors must be L2-normalised **before** being added so that inner
|
||||
@@ -49,7 +49,7 @@ class FAISS:
|
||||
db_path: Directory for persisting the index and metadata.
|
||||
|
||||
Example:
|
||||
>>> from uniface.indexing import FAISS
|
||||
>>> from uniface.stores import FAISS
|
||||
>>> store = FAISS(embedding_size=512, db_path='./my_index')
|
||||
>>> store.add(embedding, {'person_id': '001', 'name': 'Alice'})
|
||||
>>> result, score = store.search(query_embedding)
|
||||
Reference in New Issue
Block a user