ykcheungabOrisys SDK
Orisys SDK is a Python toolkit for optical tactile sensing. It provides image capture, stitching, dense optical flow, Helmholtz-Hodge decomposition, contact/depth estimation, and several runnable examples for camera and video-based workflows.
PyPI: https://pypi.org/project/orisys/
Repository Layout
src/orisys/: installable SDK packageconfig/: default and lab/runtime JSON configurationsexamples/: runnable demos (basic_pipeline.py,mini_pipeline.py,finger_pipeline*.py) plus sharedgui/,pipeline/,viz/scripts/: maintenance, calibration, packaging, and demo scriptstools/: non-package tooling such as patch-generation utilitiestests/: automated pytest suitedocs/: usage and API documentation
Quick Start
1. Create an environment
conda create -n orisys python=3.10
conda activate orisys
2. Install the SDK in editable mode
pip install -e .
Optional GPU extras:
pip install -e .[cuda12]
or
pip install -e .[cuda13]
3. Run tests
pytest
4. Run an example
ORY-BOX:
python examples/basic_pipeline.py -v 0 -c box
ORY-MINI:
python examples/mini_pipeline.py -v 0 -c mini
ORY-FINGER:
python examples/finger_pipeline.py --camera_id 0 --config finger
Run commands from the repository root.
Configuration
The canonical configuration files now live in the repo-level config/ folder.
The package loader orisys.configs resolves configs from:
- an explicit JSON file path
- a config file stem under
config/(for examplebox,mini,finger) - a product alias such as
dd02->dd02-ov - a backward-compatible JSON
"name"match, such as"prototype"
Examples:
import orisys
sensor = orisys.Sensor(0, config_name="box")
cfg = orisys.export_config("box", "./config/exported_config.json")
Examples and Tools
- Example entry points:
examples/README.md - User documentation:
docs/get_started.md - Build launcher:
python build/run.py(seebuild/README.md) - Manylinux / PyPI binary wheel:
docs/publishing.md - Specialized utilities:
tools/
Publishing (Linux manylinux)
To build PyPI-compliant CPython 3.10/3.11/3.12 Linux wheels from Windows Docker Desktop:
python build/run.py manylinux
Upload candidates land in dist/repaired/. Full TestPyPI → PyPI steps:
docs/publishing.md.
Notes for Collaborators
- Generated build outputs such as
build/,dist/, andSDK/are intentionally ignored. - Large or generated tooling outputs should stay under ignored output folders rather than be committed.
- If you add new images or arrays that should be versioned, review
.gitignorefirst. - Never commit PyPI / TestPyPI API tokens.