Skip to content
pubscout

Compare

The strongest value in each row is marked.

Attributegoogle_mlkit_object_detection

Detect and track objects in images or live camera feeds using Google's ML Kit.

pub.devRemove
hand_landmarker

Real-time hand landmark detection on Android using Google MediaPipe and JNI.

pub.devRemove
ScoresHealth87 (best)54
Maintenance98 (best)66
Quality9397 (best)
Popularity77 (best)20
Community80 (best)6
Compatibility75 (best)67
UsageDownloads (30 days)16k (best)806
Likes92 (best)11
Pub points150 / 160160 / 160 (best)
GitHub stars1.3k (best)5
MaintenanceLatest release1.0.3, 9 days ago (best)3.1.0, 2 months ago
Major versions in 12 months11
Contributors59 (best)1
Publisherflutter-ml.devUnverified
FitPlatformsAndroid, iOSAndroid
WASM-readyNoNo
KindPlugin (native code)Plugin (native code)
Code generationNoNo
Direct dependencies1 (best)4
LicenseMITMIT
Setup effortHighLow
JudgementStrengths
  • Supports base model, local custom models, and Firebase models.
  • Handles both single images and live camera streams via DetectionMode.
  • Non-blocking live tracking on a background thread
  • Configurable GPU or CPU delegate
  • Bundled MediaPipe hand_landmarker.task model requires no manual setup
  • Optimized JNI bridge with custom YUV to ARGB conversion
Weaknesses
  • Mobile-only (Android and iOS); web and other platforms are not supported.
  • Requires manual platform configuration including specific iOS deployment targets and Xcode build settings.
  • Limited to the Android platform
  • Requires minimum Flutter SDK 3.44 and JDK 17
Warnings
  • Native setup
  • Native setup
google_mlkit_object_detection vs hand_landmarker: which Flutter package should you use? | pubscout