ML kit camera text recognition not working when navigated from another screen

Question

I asked a previous question here where the camera preview screen with ocr wouldn't work when navigated from another screen and I managed to get it to work but now the problem is the ocr function doesn't work at all. By that I mean it's not detecting any text. Also when run as is the camera preview would take the entire screen but when navigated from another screen, like a homescreen it only takes the bottom half of the screen.

@Composable
fun CameraPreviewScreen() {
    val lensFacing = CameraSelector.LENS_FACING_BACK
    val lifecycleOwner: LifecycleOwner = LocalLifecycleOwner.current
    val context = LocalContext.current
    val cameraController: LifecycleCameraController =
        remember { LifecycleCameraController(context) }
    var detectedText: String by remember { mutableStateOf("Recognized text goes here") }

    fun onTextUpdated(updatedText: String) {
        detectedText = updatedText

    }

    AndroidView(
        factory = {
            val previewView = PreviewView(context).apply {
                this.scaleType = scaleType
                layoutParams = ViewGroup.LayoutParams(
                    ViewGroup.LayoutParams.MATCH_PARENT,
                    ViewGroup.LayoutParams.MATCH_PARENT
                )
                implementationMode = PreviewView.ImplementationMode.COMPATIBLE
            }
            val cameraProviderFuture = ProcessCameraProvider.getInstance(context)
            cameraProviderFuture.addListener({
                val cameraProvider = cameraProviderFuture.get()

                // Preview
                val preview = Preview.Builder()
                    .build()
                    .also {
                        it.setSurfaceProvider(previewView.surfaceProvider)
                    }

                try {
                    // Must unbind the use-cases before rebinding them.
                    cameraProvider.unbindAll()
                    cameraController.bindToLifecycle(lifecycleOwner)
                    cameraProvider.bindToLifecycle(
                        lifecycleOwner, cameraSelector, preview
                    )
                } catch (exc: Exception) {
                    Log.e(TAG, "Use case binding failed", exc)
                }
            }, ContextCompat.getMainExecutor(context))
            previewView.also { previewView ->
                initializeTextRecognition(
                    context = context,
                    cameraController = cameraController,
                    lifecycleOwner = lifecycleOwner,
                    previewView = previewView,
                    onDetectedTextUpdated = ::onTextUpdated,
                )
            }
        },
        modifier = Modifier.fillMaxSize(),
    )

    Text(
        modifier = Modifier
            .fillMaxWidth()
            .background(androidx.compose.ui.graphics.Color.White)
            .padding(16.dp),
        color = Color.Black,
        text = detectedText,
    )
}

//text recognition function
private fun initializeTextRecognition(
    context: Context,
    cameraController: LifecycleCameraController,
    lifecycleOwner: LifecycleOwner,
    previewView: PreviewView,
    onDetectedTextUpdated: (String) -> Unit,
) {
    cameraController.imageAnalysisTargetSize = CameraController.OutputSize(AspectRatio.RATIO_16_9)
    cameraController.setImageAnalysisAnalyzer(
        ContextCompat.getMainExecutor(context),
        TextRecognitionAnalyzer(onDetectedTextUpdated = onDetectedTextUpdated)
    )
    cameraController.bindToLifecycle(lifecycleOwner)
    previewView.controller = cameraController
}

I suspect that the problem lies with the text recognition function not binding to the previewview or such correctly because it runs properly when the composable is ran directly in the main activity.