MediaPipe FaceDetector Module POC Error - Error: StartGraph failed: $Service "kGpuService", required by node

[aaronxiongcday]

Have I written custom code (as opposed to using a stock example script provided in MediaPipe)

Yes

OS Platform and Distribution

Android OS 9

MediaPipe Tasks SDK version

asks-vision@0.10.0

Task name (e.g. Image classification, Gesture recognition etc.)

FaceDetector module initialization

Programming Language and version (e.g. C++, Python, Java)

HTML, JavaScript

Describe the actual behavior

The web app fails at initializing the FaceDetector module

Describe the expected behaviour

The web app successfully initializes the FaceDetector module

Standalone code/steps you may have used to try to get what you need

Implementation:
const initializefaceDetector = async () => {
  try {
    const vision = await FilesetResolver.forVisionTasks(
      "https://cdn.jsdelivr.net/npm/@mediapipe/tasks-vision@0.10.0/wasm"
    );
    faceDetector = await FaceDetector.createFromOptions(vision, {//line 31
      baseOptions: {
        //modelAssetPath: `https://storage.googleapis.com/mediapipe-models/face_detector/blaze_face_short_range/float16/1/blaze_face_short_range.tflite`,
        modelAssetPath: "./blaze_face_short_range.tflite",
        delegate: "GPU"
      },
      runningMode: runningMode
    });
    demosSection.classList.remove("invisible");
  } catch (error) {
    console.error("Error initializing face detector:", error);
  }
};
initializefaceDetector();

The modified MediaPipe FaceDetector demo web-app is hosted by the GitHub pages feature, and the web-app is launched on the Android 9 WebView browser (version 96.0.4664.141).

Getting the runtime console error:
Error initializing face detector: Error: StartGraph failed: $Service "kGpuService", required by node mediapipe_tasks_vision_face_detector_facedetectorgraph__mediapipe_tasks_components_processors_imagepreprocessinggraph__ImageCloneCalculator, was not provided and cannot be created: emscripten_webgl_create_context() returned error 0; StartRun failed
    at u.handleErrors (tasks-vision@0.10.0:7)
    at u.setGraph (tasks-vision@0.10.0:7)
    at u.refreshGraph (tasks-vision@0.10.0:7)
    at tasks-vision@0.10.0:7
    at async createTaskRunner (tasks-vision@0.10.0:7)
    at async initializefaceDetector (libs.js:31)

Other info / Complete Logs

No response

[kuaashish]

Hi @aaronxiongcday,

This error often occurs when attempting to use the GPU on a device that lacks GPU support. Could you please verify if your device is GPU-compatible and keep us informed about the issue?

Thank you!!

[aaronxiongcday]

Hi @kuaashish,
I have tried the following, switched to use CPU, but still getting the same error:

faceDetector = await FaceDetector.createFromOptions(vision, {
baseOptions: {
modelAssetPath: https://storage.googleapis.com/mediapipe-models/face_detector/blaze_face_short_range/float16/1/blaze_face_short_range.tflite,
delegate: "CPU"
},
runningMode: runningMode
});

Other than the CPU and GPU options, is there third option?

[kuaashish]

Hi @aaronxiongcday,

To help us diagnose the error you are encountering, could you confirm if you are testing this scenario on an emulator or a physical device? We have identified that the issue is likely to occur on emulators due to limitations in emulated environments. If you are using an emulator, running the test on a physical device would be helpful. Please let us know the outcome after running the test on a physical device.

Thank you!!

[aaronxiongcday]

Hi @kuaashish,
The scenario is testing on a physical device, which the Android OS is 9, and the application is launched via Webview browser.

[kuaashish]

Hi @aaronxiongcday,

Could you please test it with the latest version 0.10.13 once? We always recommend using the latest version as it includes new features and fixes for bugs.

Thank you!!

[github-actions]

This issue has been marked stale because it has no recent activity since 7 days. It will be closed if no further activity occurs. Thank you.

[github-actions]

This issue was closed due to lack of activity after being marked stale for past 7 days.

[google-ml-butler]

Are you satisfied with the resolution of your issue?
Yes
No