Object Detection App not working (Efficient net lite)

General Discussion
, ,
#1

I have trained a object detection model using ssd_mobilenet(fine-tuned) architecture and used it in my android application and it runs good.
but when i use tflite-model-maker to train a new custom object detection model using efficient net lite0 architecture and I use it in my same application it gives the following error.

Fatal signal 11 (SIGSEGV), code 1, fault addr 0x4 in tid 30204 (pool-1-thread-1) error in android native

can’t figure out what is the problem here.

#2

Update to latest tensorflow lite task vision dependency in your gradle file.

implementation ‘org.tensorflow:tensorflow-lite-task-vision:0.4.1’

#4

Your app is crashing due to dereferencing a null pointer.
Perhaps you are not checking for null return values from TF Lite function calls?
There’s not really enough info in your message to debug this further, but web search for “how do I get a stack trace for a SIGSEGV in Android native code” should help you to figure out how to debug this further.