mediapipe

mediapipe copied to clipboard

Please make sure that this is a solution issue.

System information (Please provide as much relevant information as possible)

• Have I written custom code (as opposed to using a stock example script provided in Mediapipe):
  1. My code based on your example [mediapipe/examples/android/src/java/com/google/mediapipe/]
  2. I add addMultiStreamCallback to check performance as :

             processor.getGraph().addMultiStreamCallback(
                OUTPUT_LANDMARKS_STREAM,
                packets -> {
                    Packet packet = packets.get(0);

                    long currTimeMs = System.currentTimeMillis();
                    float fps = 1000 / (currTimeMs - oldTimeMs);

                    float avr = getFpsAverage(fps);

                    if (packet.isEmpty()) {
                        Log.i(TAG, "[0] report rate : " + fps + ", " + avr);
                    } else {
                        Log.i(TAG, "[1] report rate : " + fps + ", " + avr);
                    }

                    oldTimeMs = currTimeMs;
                });

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04, Android 11, iOS 14.4): Linux Ubuntu 22.04

• MediaPipe version: v0.8.10.2

• Bazel version: bazel 5.2.0

• Solution (e.g. FaceMesh, Pose, Holistic): Holistic

• Programming Language and version ( e.g. C++, Python, Java): android (c++, Java)

Describe the expected behavior: I expected It worked over 20 fps on device, but It worked under 10 fps when all landmarks(face, body pose, hand) were detected. it was mentioned your blog : https://ai.googleblog.com/2020/12/mediapipe-holistic-simultaneous-face.html My chipset is qcs8250 (I think my device is better than S9+)

:56:49.479 7373-7430/com.lge.myholistic I/MainActivity: [1] report rate : 7.0, 7.4 2022-07-29 10:56:49.608 7373-7426/com.lge.myholistic I/MainActivity: [1] report rate : 7.0, 7.4 2022-07-29 10:56:49.751 7373-7429/com.lge.myholistic I/MainActivity: [1] report rate : 6.0, 7.1 2022-07-29 10:56:49.878 7373-7429/com.lge.myholistic I/MainActivity: [1] report rate : 7.0, 7.1 2022-07-29 10:56:50.002 7373-7431/com.lge.myholistic I/MainActivity: [1] report rate : 8.0, 7.2 2022-07-29 10:56:50.137 7373-7430/com.lge.myholistic I/MainActivity: [1] report rate : 7.0, 7.2 2022-07-29 10:56:50.253 7373-7431/com.lge.myholistic I/MainActivity: [1] report rate : 8.0, 7.2 2022-07-29 10:56:50.365 7373-7429/com.lge.myholistic I/MainActivity: [1] report rate : 9.0, 7.5 2022-07-29 10:56:50.488 7373-7426/com.lge.myholistic I/MainActivity: [1] report rate : 8.0, 7.5 2022-07-29 10:56:50.615 7373-7431/com.lge.myholistic I/MainActivity: [1] report rate : 7.0, 7.4 2022-07-29 10:56:50.740 7373-7432/com.lge.myholistic I/MainActivity: [1] report rate : 8.0, 7.5 2022-07-29 10:56:50.865 7373-7432/com.lge.myholistic I/MainActivity: [1] report rate : 7.0, 7.5 2022-07-29 10:56:50.985 7373-7433/com.lge.myholistic I/MainActivity: [1] report rate : 8.0, 7.7

Standalone code you may have used to try to get what you need :

If there is a problem, provide a reproducible test case that is the bare minimum necessary to generate the problem. If possible, please share a link to Colab/repo link /any notebook:

Other info / Complete Logs : Include any logs or source code that would be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached:

addtionally,

I wrote BUILD file to make aar:

load(
    "//mediapipe/java/com/google/mediapipe:mediapipe_aar.bzl",
    "mediapipe_aar"
)

mediapipe_aar(
    name = "mediapipe_holistic_lge",
    calculators = ["//mediapipe/graphs/holistic_tracking:holistic_tracking_gpu_deps"],
)

Hi @ar0127 , Do you have still issue w.r.t this issue?

Hi @sureshdagooglecom

Yes, I have this issue still.

I built aar and I used it with your holistic graph with no modification. I just followed your guide.

I want to get your opion to improve performance.

thanks

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

Hi @sureshdagooglecom

I reopened the issue. I didn't know the the issue closed when i wrote comment.

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you.

Closing as stale. Please reopen if you'd like to work on this further.

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

Hi @ar0127 , We have closed issue running on thread , could you please check and verify https://github.com/google/mediapipe/issues/2465

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you.

Closing as stale. Please reopen if you'd like to work on this further.

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