The sample_camera_gmsl is able to write a h264 stream to disk. How can I output the h264 stream to network?
Dear yc563,
If use object detect sample, please help to check --help option like below and you can use the captured clip in video option part. Thanks.
nvidia@nvidia:/usr/local/driveworks-0.6/bin$./sample_object_detector --help
–tensorRT_model: default=…/./data/samples/detector/tensorRT_model.bin
–video: default=…/./data/samples/sfm/triangulation/video_0.h264
–yuv2rgb: default=cuda
Dear SteveNV,
I want to use the serializer to do a function callback to retrieve a h264 frame when a camera image is received.
As a csiPort can have up to 4 sibling cameras, do I need to run dwSensorSerializer_initialize for each sibling in a csiPort?
Dear yc563,
You mean you need to control each port in one camera group? Thanks.
Dear SteveNV,
Yes, i need to extract data (using a callback) from each port in one camera group.
If I have 4 ports in each camera group, do i need to run dwSensorSerializer_initialize and dwSensorSerializer_start four times?
I solved it by calling a dwSensorSerializer for each port. Thanks.
yc563,
So, you just used something else for the video-file? Can you give a snippet on how you stream to the network? Thanks!
1 Like