Cortex
Copyright (c) 2026 Steven Wang

This product includes software developed by third parties, listed below
with their original licenses. Source-code dependencies are pinned in
`cortex/pyproject.toml` and `cortex/apps/browser_extension/package.json`.

---

## Bundled binary models

cortex/models/face_landmarker.task
    MediaPipe Face Landmarker model bundle from Google LLC.
    Licensed under the Apache License, Version 2.0.
    https://www.apache.org/licenses/LICENSE-2.0
    Source: https://developers.google.com/mediapipe/solutions/vision/face_landmarker

cortex/models/tscan.onnx  (optional, user-supplied at runtime — NOT bundled)
    TS-CAN remote-photoplethysmography model.
    If you provide this file, ensure your local copy complies with the
    license of the checkpoint you exported from. The Cortex repository
    does not ship pre-trained TS-CAN weights.
    Reference: Liu, X., et al. (2020). "Multi-Task Temporal Shift
    Attention Networks for On-Device Contactless Vitals Measurement."

## Major runtime dependencies (see pyproject.toml for full list)

MediaPipe                Apache-2.0    https://github.com/google/mediapipe
OpenCV                   Apache-2.0    https://github.com/opencv/opencv
ONNX Runtime             MIT           https://github.com/microsoft/onnxruntime
NumPy                    BSD-3-Clause  https://numpy.org/
SciPy                    BSD-3-Clause  https://scipy.org/
PySide6                  LGPL-3.0      https://wiki.qt.io/Qt_for_Python
FastAPI                  MIT           https://github.com/tiangolo/fastapi
Uvicorn                  BSD-3-Clause  https://github.com/encode/uvicorn
websockets               BSD-3-Clause  https://github.com/python-websockets/websockets
pynput                   LGPL-3.0      https://github.com/moses-palmer/pynput
Pydantic                 MIT           https://github.com/pydantic/pydantic
structlog                Apache-2.0    https://github.com/hynek/structlog
keyring                  MIT           https://github.com/jaraco/keyring
Anthropic SDK (Python)   MIT           https://github.com/anthropics/anthropic-sdk-python

## Browser extension

Plasmo                   MIT           https://github.com/PlasmoHQ/plasmo
React                    MIT           https://github.com/facebook/react

## Scientific references

The rPPG pipeline implements published algorithms; see in-code
docstrings for citations:

  - Wang, W., den Brinker, A. C., Stuijk, S., & de Haan, G. (2017).
    "Algorithmic principles of remote PPG." IEEE TBME.
  - de Haan, G., & Jeanne, V. (2013).
    "Robust pulse rate from chrominance-based rPPG." IEEE TBME.

This NOTICE file is informational. The MIT LICENSE in this repository
applies to Cortex's own source code; third-party components remain
under their respective licenses.