OPT8241-CDK-EVM: voxelsdl-example DepthCapture capture data format and file structure?

Finley Huang

Part Number: OPT8241-CDK-EVM

Dear Sir,

1. I had run voxelsdl-example's DepthCapture, when I used "pointcloud" type
capture data from opt8241, it's size was 1228812 bytes, but 320*240*16 = 1228800 bytes,
more than 12 bytes, is header? what is header structure?
Do other types of files has the same header?

2. when I used "-n 1 -t depth" to capture image data, Is the depth map file content
the Z-axis value of pointcloud data (X, Y, Z, I) and why is its size 614412?
If the depth file is a Z-axis value, the size of 1 frame should be 307200 bytes.

over 6 years ago

0 Suramya Gupta over 6 years ago

TI__Expert 3975 points

Hi Finley,

As you can see from the DepthCapture.cpp code, each of the frame types is also saving the frame id as well as the timestamp. This is taking 12 bytes.
The same header is saved for all frame types.
For the depth data, two kinds of data are saved, the depth (Z axis from the point cloud data), as well as the intensity (amplitude normalized between 0 and 1). See the code snippet from https://github.com/3dtof/voxelsdk-examples/blob/master/DepthCapture/DepthCapture.cpp#L287

So, for depth capture, you'll see (4*240*320 + 4*240*320 + 12) bytes for a frame.

The header structure is:

4 bytes for id and 8 bytes for timestamp.

Suramya

0 Finley Huang over 6 years ago in reply to Suramya Gupta

Prodigy 140 points

Dear Suramya,
This resolved my question.
Thanks you.

Sensors

Sensors forum

OPT8241-CDK-EVM: voxelsdl-example DepthCapture capture data format and file structure?