v0.4.1

What's Changed

• fix: broken protobuf related dependencies by @VLucet in #28

Full Changelog: v0.4.0...v0.4.1

v0.4.0

What's Changed

• Made ImageDataset name optional by @jaivardhan-bhola in #20
• Add IPTC keyword tagging parsing by @bencevans in #25
• Speed up Image Enumeration by @bencevans in #17

New Contributors

• @jaivardhan-bhola made their first contribution in #20

Full Changelog: v0.3.0...v0.4.0

v0.3.0

What's Changed

• Update Python Versions on Actions by @bencevans in #12
• Run CodeQL Analysis on Actions by @bencevans in #8
• Add a timeout for loading images from HTTP/S by @bencevans in #13

Full Changelog: v0.2.5...v0.3.0

v0.2.5

What's Changed

• Fix fast exif where number of paths less than batch size by @bencevans in #10
• Test README Notebook on Actions by @bencevans in #11

Full Changelog: v0.2.3...v0.2.5

v0.2.3

Fixed a bug with a missed renaming of function.

Full Changelog: v0.2.2...v0.2.3

v0.2.2

Removes declarative imports from init.py files, so the rest of the tree is only imported as and when required.

v0.2.0

Broadened dependency ranges, including TensorFlow, to increase compatibility with workspaces.

v0.1.0

The initial release of CamTrapML.

Features Including:

• Loading of Camera Trap Imagery Datasets
• Loading images from both local file systems and HTTP/HTTPS
• Extracting EXIF data from image sets, often containing date/time, sequence or tagging metadata
• Detection of animal, human and vehicle classes using the Microsoft MegaDetector
• Detection using custom Tensorflow Object Detection API models
• Extracting and removing detections from images. Particularly useful for removing humans from imagery or extracting animals for further classification.