huawei_ml_image 3.13.0+301

Huawei ML Kit Image Plugin for Flutter. Provides capabilities like image classification, object detection and image super resolution.

3.13.0+301

• Android build modernized for Gradle 8 / AGP 8
  • Updated the Android build to AGP 8.3.2 (Gradle wrapper 8.4), added the plugin namespace (com.huawei.hms.flutter.mlimage), and moved to Java 17 / compileSdk 34.
  • Updated the example app to Gradle 8.13, AGP 8.11.1, Kotlin 2.2.20, and Java 17.
• Image segmentation stays fully offline (bundled models kept)
  • The bundled image-segmentation model packs (multiclass / body / hair, all 3.7.0.302, matched to the segmentation lib) are retained, so segmentation works offline with no runtime download.
  • These model packs share the namespace com.huawei.hms.mlkit.imgseg; AGP 9 rejects this duplicate namespace with no bypass flag. This package therefore targets Gradle 8 / AGP 8 host apps. On AGP 8, host apps must set android.uniquePackageNames=false in gradle.properties.
• Dependency updates
  • Bumped ml-computer-vision-cloud to 3.18.6.305 (the previous 3.13.1.302 pulled a removed transitive dependency).
  • Bumped mindspore-lite to 5.1.1.300.

3.13.0+300

• Breaking Change: Migrated to declarative gradle.

3.12.0+300

• Dart improvements.

3.11.0+300

• Updated targetSdkVersion to 31, to make sure that your app can run properly on Android 12.

3.7.0+300

• Added setUserRegion and getCountryCode methods to MLImageApplication.

• Breaking Changes:

  • With this release, MLImagePermissions has been removed. You are expected to handle required permissions on your own. You can learn more about the required permissions from our official documentations.
  • Modified the internal structure of the plugin. Please use import package:huawei_ml_image/huawei_ml_image.dart not to get any errors.

3.2.0+301

• Deleted the capability of prompting users to install HMS Core (APK).

3.2.0+300

• Initial release.