feat: facial recognition (#2180)

machine-learning/.dockerignore

@@ -1 +1,3 @@
-venv/
+venv/
+*.zip
+*.onnx

machine-learning/.gitignore

@@ -3,4 +3,170 @@
 upload/
 venv/
 __pycache__/
-model-cache/
+model-cache/
+
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+.idea/
+
+
+*.onnx
+*.zip

machine-learning/Dockerfile

@@ -8,7 +8,8 @@ RUN python -m venv /opt/venv
 RUN /opt/venv/bin/pip install torch --index-url https://download.pytorch.org/whl/cpu
 RUN /opt/venv/bin/pip install transformers tqdm numpy scikit-learn scipy nltk sentencepiece fastapi Pillow uvicorn[standard]
 RUN /opt/venv/bin/pip install --no-deps sentence-transformers
-
+# Facial Recognition Stuff
+RUN /opt/venv/bin/pip install insightface onnxruntime
 
 FROM python:3.10-slim
 

machine-learning/src/main.py

@@ -1,9 +1,13 @@
+import os
+import numpy as np
+import cv2 as cv
+import uvicorn
+
+from insightface.app import FaceAnalysis
 from transformers import pipeline
 from sentence_transformers import SentenceTransformer, util
 from PIL import Image
 from fastapi import FastAPI
-import uvicorn
-import os
 from pydantic import BaseModel
 
 
@@ -15,15 +19,6 @@ class ClipRequestBody(BaseModel):
     text: str
 
 
-is_dev = os.getenv('NODE_ENV') == 'development'
-server_port = os.getenv('MACHINE_LEARNING_PORT', 3003)
-server_host = os.getenv('MACHINE_LEARNING_HOST', '0.0.0.0')
-
-app = FastAPI()
-
-"""
-Model Initialization
-"""
 classification_model = os.getenv(
     'MACHINE_LEARNING_CLASSIFICATION_MODEL', 'microsoft/resnet-50')
 object_model = os.getenv('MACHINE_LEARNING_OBJECT_MODEL', 'hustvl/yolos-tiny')
@@ -31,9 +26,15 @@ clip_image_model = os.getenv(
     'MACHINE_LEARNING_CLIP_IMAGE_MODEL', 'clip-ViT-B-32')
 clip_text_model = os.getenv(
     'MACHINE_LEARNING_CLIP_TEXT_MODEL', 'clip-ViT-B-32')
+facial_recognition_model = os.getenv(
+    'MACHINE_LEARNING_FACIAL_RECOGNITION_MODEL', 'buffalo_l')
+
+cache_folder = os.getenv('MACHINE_LEARNING_CACHE_FOLDER', '/cache')
 
 _model_cache = {}
 
+app = FastAPI()
+
 
 @app.get("/")
 async def root():
@@ -73,6 +74,36 @@ def clip_encode_text(payload: ClipRequestBody):
     return model.encode(text).tolist()
 
 
+@app.post("/facial-recognition/detect-faces", status_code=200)
+def facial_recognition(payload: MlRequestBody):
+    model = _get_model(facial_recognition_model, 'facial-recognition')
+    assetPath = payload.thumbnailPath
+    img = cv.imread(assetPath)
+    height, width, _ = img.shape
+    results = []
+    faces = model.get(img)
+    for face in faces:
+        if face.det_score < 0.7:
+            continue
+        x1, y1, x2, y2 = face.bbox
+        # min face size as percent of original image
+        # if (x2 - x1) / width < 0.03 or (y2 - y1) / height < 0.05:
+        #     continue
+        results.append({
+            "imageWidth": width,
+            "imageHeight": height,
+            "boundingBox": {
+                "x1": round(x1),
+                "y1": round(y1),
+                "x2": round(x2),
+                "y2": round(y2),
+            },
+            "score": face.det_score.item(),
+            "embedding": face.normed_embedding.tolist()
+        })
+    return results
+
+
 def run_engine(engine, path):
     result = []
     predictions = engine(path)
@@ -93,12 +124,22 @@ def _get_model(model, task=None):
     key = '|'.join([model, str(task)])
     if key not in _model_cache:
         if task:
-            _model_cache[key] = pipeline(model=model, task=task)
+            if task == 'facial-recognition':
+                face_model = FaceAnalysis(
+                    name=model, root=cache_folder, allowed_modules=["detection", "recognition"])
+                face_model.prepare(ctx_id=0, det_size=(640, 640))
+                _model_cache[key] = face_model
+            else:
+                _model_cache[key] = pipeline(model=model, task=task)
         else:
-            _model_cache[key] = SentenceTransformer(model)
+            _model_cache[key] = SentenceTransformer(
+                model, cache_folder=cache_folder)
     return _model_cache[key]
 
 
 if __name__ == "__main__":
-    uvicorn.run("main:app", host=server_host,
-                port=int(server_port), reload=is_dev, workers=1)
+    host = os.getenv('MACHINE_LEARNING_HOST', '0.0.0.0')
+    port = int(os.getenv('MACHINE_LEARNING_PORT', 3003))
+    is_dev = os.getenv('NODE_ENV') == 'development'
+
+    uvicorn.run("main:app", host=host, port=port, reload=is_dev, workers=1)