TDA4VM: Embedded data

Hi Raj,

Did not get your question, do you mean to extract embedded information and process it, after VISS node? VISS outputs in YUV formats, which does not include these information. If this data is part of the frame, it should be processed after the CSIRX node, since this node outputs raw image with embedded data.. 

VISS node just skips embedded data and processes actual RAW data and outputs in YUV format.

Regards,

Brijesh

Yes, extract embedded information after the viss node. Why does the VISS node just skip embedded data? Why restrict output to only output image data?

We are getting good quality images after the VISS node. We need embedded data in the VISS node. How to do that?

Because VISS is a pixel processing module and it is going to change the contents of the image data. If you want embedded data also to be processed in VISS, set the size of the image as full in VISS. 

If embedded data is having same data type as image data, then it will be part of the same image and you could extract it from the image buffer even in viss node.

Hi Brijesh,

Please explain in the details given below function which parameter the size of the image must be set to full  

Regards

Raj

Hi Brijsth

Given below function which parameter the size of the image must be set to full ( explain in details)  

VX_API_CALL tivxVpacVissNode (vx_graph graph, vx_user_data_object configuration,vx_user_data_object ae_awb_result,
vx_user_data_object dcc_buf,
tivx_raw_image raw,
vx_image output0,
vx_image output1,
vx_image output2,
vx_image output3,
vx_image output4,
vx_user_data_object h3a_output,
vx_distribution histogram
)

Regards

Raj

Hi Raj Musthu,

I would suggest you to go through the existing documentation of OpenVX and single/multi camera demo to understand these parameters.

Since this question is not related to embedded data, i am closing this ticket. Please start new ticket for new question.

Regards,

Brijesh

This question related to embedded data  . You have mentioned to set the image full size in VISSS node

The last question was about VISS parameter details, it is not about embedded data. 

You have mentioned that (for my query).

Because VISS is a pixel processing module, it is going to change the contents of the image data. If you want embedded data to also be processed in VISS, set the size of the image to "full" in VISS.

How to set the size of the image as full in the VISS node ( which parameter of the viss node )

check the IssSensor_CreateParams parameters, include meta size in the image size.