TFLite输入格式

Question

嗨，我想根据github上的tflite示例了解tflites输入格式的工作方式。为什么仍然要使用这种格式。原因是我想在没有camera 2 api的情况下实现此功能，但与python不同，它不是那么容易操作或查看图像

 @Override
protected void onPreviewSizeChosen(Size size, int rotation) {
    int cropSize = TF_OD_API_INPUT_SIZE;
    initTFLite(cropSize);
    Log.d("PREVIEW", size.toString());

    previewHeight = size.getHeight();
    previewWidth = size.getWidth();

    sensorOrientation = rotation - getScreenOrientation();

    // Create empty bitmap
    rgbFrameBitmap = Bitmap.createBitmap(previewWidth, previewHeight, Bitmap.Config.ARGB_8888);
    croppedBitmap = Bitmap.createBitmap(cropSize, cropSize, Bitmap.Config.ARGB_8888);

    // I dont understand what this does
    frameToCropTransform = ImageUtils.getTransformationMatrix(
            previewWidth, previewHeight,
            cropSize, cropSize,
            sensorOrientation, MAINTAIN_ASPECT
    );

    cropToFrameTransform = new Matrix();
    frameToCropTransform.invert(cropToFrameTransform);
}

@Override
protected void processImage() {
    // No mutex needed as this method is not reentrant.
    if (computingDetection) {
        readyForNextImage();
        return;
    }

    computingDetection = true;
    // Prepare image
    // What does this set pixels do anyway
    rgbFrameBitmap.setPixels(getRgbBytes(), 0, previewWidth, 0, 0, previewWidth, previewHeight);
    readyForNextImage();

    final Canvas canvas = new Canvas(croppedBitmap);
    canvas.drawBitmap(rgbFrameBitmap, frameToCropTransform, null);
    // For examining the actual TF input.
    if (SAVE_PREVIEW_BITMAP) {
        ImageUtils.saveBitmap(croppedBitmap);
    }

    runInBackground(new Runnable() {
        @Override
        public void run() {
            final List<Classifier.Recognition> results = detector.recognizeImage(croppedBitmap);

            float minimumConfidence = MINIMUM_CONFIDENCE_TF_OD_API;
            switch (MODE) {
                case TF_OD_API:
                    minimumConfidence = MINIMUM_CONFIDENCE_TF_OD_API;
                    break;
            }

            Log.d("RESULTS", results.toString());

            computingDetection = false;
        }
    });