apple_vision_native 0.1.0

Flutter bindings for Apple's Vision framework: text recognition, detection, classification, segmentation and more.

0.1.0

• First publication as apple_vision_native (renamed from the unpublished workspace package apple_vision).

• Initial release: 25 modern Vision requests over Pigeon platform channels, available on iOS 27+ and macOS 27+ with older deployment targets supported.

• Text and document recognition, barcodes, faces, people and animal poses, classification, aesthetics, lens smudges, saliency, segmentation, rectangles, horizon, contours and image feature prints.

• File, Flutter asset, encoded image and BGRA pixel inputs; batch requests share one decoded image and preserve errors for individual requests.

• Normalized geometry with conversion to Flutter coordinates, raw masks with row stride and PNG encoding, and feature-print distance comparisons.

• Injectable bindings, typed errors, a Material example, Dart unit tests and integration tests against bundled images.