mediapipe
mediapipe copied to clipboard
googlesamples
Few weeks ago, I faced an issue of duplicate symbols in my iOS application when using TensorFlowLiteSwift and MediaPipeTasksVision together and application couldn't build but after the release of version...
Hi, thanks for providing a great examples. It is amazing indeed. While I am deploying the test app for LLM chat, I tried two models which are Gemma and phi-2....
I am using the[ image generation sample](https://github.com/googlesamples/mediapipe/tree/main/examples/image_generation) and tried running on my android device Below is the error ``` FATAL EXCEPTION: DefaultDispatcher-worker-1 Process: com.google.mediapipe.examples.imagegeneration, PID: 29768 java.lang.UnsatisfiedLinkError: dlopen failed: library...
 The example labeled as "Mixed" here, is actually 100% Chinese. https://github.com/googlesamples/mediapipe/blame/main/examples/language_detector/python/%5BMediaPipe_Python_Tasks%5D_Language_Detector.ipynb
Just checking out the current master: 0fc6e2b809b13b7cb1de1792288409624f5f6007 And trying to run it on my Samsung Galaxy S7 Tablet, gives me the following error: ``` 2024-03-05 17:17:56.865 23156-23239 native com...apipe.examples.poselandmarker I...
`ld: warning: ignoring duplicate libraries: '-lc++' duplicate symbol '_OBJC_IVAR_$_MinMaxLocResult._minLoc' in: /Volumes/Samsung/practice/MeasurementDemo/Pods/MediaPipeTasksCommon/frameworks/graph_libraries/libMediaPipeTasksCommon_device_graph.a[2034](MinMaxLocResult.o) /Volumes/Samsung/practice/MeasurementDemo/MeasurementDemo/opencv2.framework/Versions/A/opencv2[arm64][884](MinMaxLocResult.o)` Our project uses opencv2 to generate conflicts.
Pose landmark on Python crashes on version mediapipe==0.10.10 `https://github.com/googlesamples/mediapipe/blob/0fc6e2b809b13b7cb1de1792288409624f5f6007/examples/pose_landmarker/python/%5BMediaPipe_Python_Tasks%5D_Pose_Landmarker.ipynb#L221` ` "detector = vision.PoseLandmarker.create_from_options(options)\n",` This line crashes saying : AttributeError: 'FieldDescriptor' object has no attribute '_default_constructor' It does run on version...
### Description Currently, in the `object_detector.ipynb` notebook, there is no straightforward way to resize images before annotations are added. This functionality is crucial for cases where the original image size...
examples/pose_landmarker/python/[MediaPipe_Python_Tasks]_Pose_Landmarker.ipynb： The google-colab package currently found can only be successfully installed in python3.7. Another package mediapipe0.10.9 needs to be successfully installed in python3.10. In this case, how should this code...
Hi, I'm trying to work with this sample project, and I want to obtain a bitmap with the roi of the image only. But when I'm getting the result in...
