How to save the output video of DNN samples in host-pc

Autonomous Vehicles DRIVE AGX Xavier DRIVE AGX Xavier General
1

Hello,

I would like to ask if there is an automatic way of saving the output of dnn samples provided by NVIDIA DRIVE. For example, i run DriveNet, SignNet etc and everything works fine. However, i need to save the output video with the bounding boxes, the annotations, etc in order to use it for my research.
Should i modify the .cpp file writing my own code in c++? Does NVIDIA Drive provide a library or a special module to export and save the processed videos?

Thank you in advance.

Please provide the following info (check/uncheck the boxes after creating this topic):
Software Version
DRIVE OS Linux 5.2.6
DRIVE OS Linux 5.2.6 and DriveWorks 4.0
DRIVE OS Linux 5.2.0
DRIVE OS Linux 5.2.0 and DriveWorks 3.5
NVIDIA DRIVEā„¢ Software 10.0 (Linux)
NVIDIA DRIVEā„¢ Software 9.0 (Linux)
other DRIVE OS version
other

Target Operating System
[+] Linux
QNX
other

Hardware Platform
NVIDIA DRIVEā„¢ AGX Xavier DevKit (E3550)
NVIDIA DRIVEā„¢ AGX Pegasus DevKit (E3550)
other

SDK Manager Version
1.7.1.8928
other

Host Machine Version
[+] native Ubuntu 18.04
other

1 Like
2

May I know why you need to save the output? Can you use HDMI recorder or screen recorder tool?

3

Firstly,

I would like to thank you for your immediate response. I need to save the output for DNN transfer learning for my research. Thus, I need the output of a video that involves the annotations and the bounding boxes. Recording my screen will not provide this information.

I would be grateful, if you could inform me about the way I can save this output (For example, how i could modify the main.cpp code file).

Thank you in advance

4

Dear @ipapadea,
So the ask is you want a bounding box annoataed video output which can be repurposed/replayed later?

5

Dear @SivaRamaKrishnaNV ,

Thank you for your response. Yes, this is what I want for my research. How this is possible?

Thank you for your time

6

Dear @ipapadea,
That is not possible with DW APIs. As we render frames and detections separately and overlay on frame. How about saving the detection for each frame, video seperately and use them later in your application ?

7

Dear @SivaRamaKrishnaNV

Thank you for your guidance and your suggestion. Is there anyway that i can save the detection for each frame, video separately with the DW APIs?

Thank you for your time

8

Dear @ipapadea,
Please check sample_camera_gmsl sample to know API usage to save video frame( it has write-file option). For saving detection output, check saving the detection output from object detector sample(the bounding box data structure rendered on frame needs to be saved) in seperate/single file for each frame and check loading these two in your application later. Please see if it works