using tflite for movenet with some error on parsing data
Hi, I would like to switch the posenet model to movenet model which is more efficient to generate the pose, however, tflite generate the following error. Initialized TensorFlow Lite runtime. flutter: Model:success Assertion failed: (TfLiteInterpreterGetOutputTensorCount(interpreter) == 4), function parsePoseNet, file /Users/tonylu/sdk/flutter/.pub-cache/hosted/pub.dartlang.org/tflite-1.1.2/ios/Classes/TflitePlugin.mm, line 1327.
- thread #1, queue = 'com.apple.main-thread', stop reason = signal SIGABRT I think this error is caused due to output format of the model is different than posenet. Is any way I can fix this issue to make movenet work with this awesome plugin? Tony
Hello tonylu, did you find a way to run movenet in flutter? If you did can you share a code snippet or example of how you did it? I have been trying to do it and searched everywhere, but it is all unsuccessful. Thanks
+1
Hello tonylu, did you find a way to run movenet in flutter? If you did can you share a code snippet or example of how you did it? I have been trying to do it and searched everywhere, but it is all unsuccessful. Thanks
still no luck sorry ....
Is this working? https://github.com/harriseldon/movenet_flutter_example
Any updates