unified_apple_vision

Description:

unified apple vision

unified_apple_vision 🍎 #



A plugin for using Apple Vision Framework with Flutter, designed to integrate multiple APIs into one plugin and process multiple analysis requests at once.
Features ⚙️ & Requirements 🧩 #
Minimum Requirements: iOS: 13.0+, macOS: 15.0+

Warning
[*] Note that this differs from the VisionAPI minimum requirements.

Status: ✅ Complete ⚠️ Problematic 👨‍💻 In Progress ❌ Not Yet



Vision API
Request
Status
iOS
macOS
description




Still Image Analysis
Classify Image
✅
13.0+
10.15+
A request to classify an image.



Generate Image Feature Print
✅
13.0+
10.15+
An image-based request to generate feature prints from an image. (FeaturePrint — vector image descriptor similar to a word vector)










Image Sequence Analysis
Generate Person Segmentation
❌
15.0+
12.0+
Produces a matte image for a person it finds in the input image.



Generate Person Instance Mask
❌
17.0+
14.0+
Produces a mask of individual people it finds in the input image.



Detect Document Segmentation
❌
15.0+
12.0+
Detects rectangular regions that contain text in the input image.










Saliency Analysis
Generate Attention Based Saliency Image
❌
13.0+
10.15+
Produces a heat map that identifies the parts of an image most likely to draw attention.



Generate Objectness Based Saliency Image
❌
13.0+
10.15+
Generates a heat map that identifies the parts of an image most likely to represent objects.










Object Tracking
Track Rectangle
⚠️
*11.0+
*10.13+
Tracks movement of a previously identified rectangular object across multiple images or video frames.



Track Object
⚠️
*11.0+
*10.13+
Tracks the movement of a previously identified object across multiple images or video frames.










Rectangle Detection
Detect Rectangle
✅
*11.0+
*10.13+
Finds projected rectangular regions in an image.










Face and Body Detection
Detect Face Capture Quality
✅
13.0+
10.15+
Produces a floating-point number that represents the capture quality of a face in a photo.



Detect Face Landmarks
✅
*11.0+
*10.13+
Finds facial features like eyes and mouth in an image.



Detect Face Rectangles
✅
*11.0+
*10.13+
Finds faces within an image.



Detect Human Rectangles
✅
13.0+
10.15+
Finds rectangular regions that contain people in an image.










Body and Hand Pose Detection
Detect Human Body Pose
❌
14.0+
11.0+
Detects a human body pose.



Detect Human Hand Pose
❌
14.0+
11.0+
Detects a human hand pose.










3D Body Pose Detection
Detect Human Body Pose 3D
❌
17.0+
14.0+
Detects points on human bodies in three-dimensional space, relative to the camera.










Animal Detection
Recognize Animals
✅
13.0+
10.15+
Recognizes animals in an image. (Now only dogs and cats are supported.)










Animal Body Pose Detection
Detect Animal Body Pose
❌
17.0+
14.0+
Detects an animal body pose.










Trajectory Detection
Detect Trajectories
❌
14.0+
11.0+
Detects the trajectories of shapes moving along a parabolic path.










Contour Detection
Detect Contours
❌
14.0+
11.0+
Detects the contours of the edges of an image.










Optical Flow
Generate Optical Flow
❌
14.0+
11.0+
Generates directional change vectors for each pixel in the targeted image.



Track Optical Flow
❌
17.0+
14.0+
Determines the direction change of vectors for each pixel from a previous to current image.










Barcode Detection
Detect Barcodes
✅
*11.0+
*10.13+
Detects barcodes in an image.










Text Detection
Detect Text Rectangles
✅
*11.0+
*10.13+
Finds regions of visible text in an image.










Text Recognition
Recognize Text
✅
13.0+
10.15+
Finds and recognizes text in an image.










Horizon Detection
Detect Horizon
❌
*11.0+
*10.13+
Determines the horizon angle in an image.










Image Alignment
Translational Image Registration
❌
*11.0+
*10.13+
Determines the affine transform necessary to align the content of two images.



Track Translational Image Registration
❌
17.0+
14.0+
An image analysis request, as a stateful request you track over time, that determines the affine transform necessary to align the content of two images.



Homographic Image Registration
❌
*11.0+
*10.13+
Determines the perspective warp matrix necessary to align the content of two images.



Track Homographic Image Registration
❌
17.0+
14.0+
An image analysis request, as a stateful request you track over time, that determines the perspective warp matrix necessary to align the content of two images.










Image Background Removal
Generate Foreground Instance Mask
❌
17.0+
14.0+
Generates an instance mask of noticable objects to separate from the background.




Important
If you have a request, please make an issue.

Install 📦 #
Add this to your pubspec.yaml:
unified_apple_vision: ^latest
copied to clipboard
Usage 🕹 #
// initialize
final vision = UnifiedAppleVision();

// create input image
final input = VisionInputImage(
bytes: image.bytes,
size: image.size,
);

// analyze
vision.analyze(
image: input,
requests: [
// add requests you wish to perform
VisionRecognizeTextRequest(
onResults: (results) {
final observations = results.ofRecognizeTextRequest; // get casted results
// some action
},
onError: (error) {
// handle error
},
),
],
);
copied to clipboard
Documents 📘 #
Contributing 🤝 #
Contributions are welcome! Please feel free to submit a Pull Request.
Adding New Request or Observation #
If you want to add a new request or observation class, you can use the tools.py script to generate the boilerplate code for you.
python3 tools.py request --name VNDetectRectanglesRequest --ios 11.0 --macos 10.13
python3 tools.py observation --name VNRectangleObservation --extends VNDetectedObjectObservation
copied to clipboard
License 📜 #
This project is licensed under the MIT License - see the LICENSE file for details.