Open
Description
Title: Memory
Describe the bug
I am using both google_mlkit_barcode_scanning: ^0.10.0
and google_mlkit_text_recognition: ^0.11.0
for a scanner app which scans both text and barcode/QR code. On my iphone 7, the app crashes with the following report. It only crashes on iphone 7, on higher devices it overheats but does not crash.
Initialized TensorFlow Lite runtime.
INFO: Initialized TensorFlow Lite runtime.
Created TensorFlow Lite XNNPACK delegate for CPU.
INFO: Created TensorFlow Lite XNNPACK delegate for CPU.
... tracking ...
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_RESOURCE (RESOURCE_TYPE_MEMORY: high watermark memory limit exceeded) (limit=1450 MB)
frame #0: 0x00000001dc3d7884 libsystem_platform.dylib`_platform_memmove + 100
libsystem_platform.dylib`:
-> 0x1dc3d7884 <+100>: stnp x8, x9, [x3]
0x1dc3d7888 <+104>: stnp x10, x11, [x3, #0x10]
0x1dc3d788c <+108>: add x3, x3, #0x20
0x1dc3d7890 <+112>: ldnp x8, x9, [x1]
Target 0: (Runner) stopped.
Lost connection to device.
the Dart compiler exited unexpectedly.
Platform (please complete the following information):
- OS: iOS
- Device: iPhone7
- OS: 17.4
- Flutter/Dart Version : 3.19.0/3.3.0
- Plugin version: ^0.10.0/^0.11.0
I have seen in previous issues like #393 some fixes were applied related to the selfie plugin, maybe something similar can be done for text recognition and barcode scanning?