The problem about app‘s interface

[kbforever24]

Can you tell me the meaning shown in the upper left corner？
Thanks.

[szaza]

Sure,
Frame: 640x480 - every frame will be resized to 640x480 - it is configurable, however it depends on the input parameters of the neural network;
View: 1080x1812 - screen resolution;
Rotation: 90 - the image is rotated 90 degree clockwise;
Inference time: 15xx ms- the time needed for the neural network to process the input;
Regards,
Zoltan

[kbforever24]

Thank you very much.
Actually,i am a newcomer to this field.I want to know why the image needs to be rotated 90 degree.
Thanks.

[szaza]

Questions are always good, so just let me know if something is not clear and I'll try to answer your questions.
Honestly, I didn't worked on this project for a while, however as I remember the image that you can see on the screen is already rotated with 90 degrees. You can check what the frame contains if you put a brake point at the right place in the code. I suppose the image will be counterclockwise.

The brakepoint should be placed somewhere around this line:

android-yolo-v2/src/org/tensorflow/yolo/view/ClassifierActivity.java

Line 97 in 0dcf513

image.close();

Check the cropped bitmap, however it can happen that it will be already rotated, because the transformation matrix is already set in the onPreviewSizeChosen() method

Best regards,
Zoltan

[kbforever24]

Thanks again for your reply.I have understood it.
Honestly, my English is poor and i hope you dont mind. I will ask you if i have other questions.

Best wishes.

[kbforever24]

Hi,I have a question about Android development.
Can you tell me about the development process of this software ?
Thanks.