Hi,
I am trying run our model using TIDL_OD usecase. I have generated inHeader_OD,inData_OD, NET_OD.BIN and PRM_OD.BIN for our network and copied to the SD card.The model is not yet trained completely.
The network takes input of size 512x512.
In the chains_tidlOD.c I have modified some parameters as follows.
#define TIDL_OD_INPUT_WIDTH (512)
#define TIDL_OD_INPUT_HEIGHT (512)
#define DEC_OUT_WIDTH (512)
#define DEC_OUT_HEIGHT (512)
For display resolution, I am maintaining the same as for default model 768x320(scaled by 2 for display).
#define TIDL_OD_DISPLAY_WIDTH (1536)
#define TIDL_OD_DISPLAY_HEIGHT (640)
When I run the model, there are no errors, but the video is black. Even if the model contains dummy parameters, I believe we should be able to see the input video(with wrong or no detection boxes). Am I right?
Do any more parameters need to be set?
Thanks in advance,
Navinprashath.R.R
