How can I capture images/video from the ZED camera for offline processing in ZED SDK/ZED tools?

The ZED SDK, ZEDfu, and other ZED tools accept three types of inputs: live images from a directly connected ZED camera, recorded video files from the ZED camera in native SVO format to perform offline processing or network streams from a remote ZED camera.

Record video files with the ZED

Open the ZED Explorer tool and hit the record button to start recording video files. The video sequence is saved with our proprietary SVO format.

SVO files contain all the required information to simulate a live camera within the ZED SDK. 

Should I use offline processing?

The advantages of the Offline Processing method are the following ones:

  • If you do not use H264 or H265 compression, no Nvidia GPU is required to record SVO files. The recording computer only needs a USB 3.0 port and an i5 or higher CPU. Most of our users use an INTEL NUC or an Nvidia Jetson board.
  • This method offers the possibility to tune the SDK parameters in order to generate different outputs from the same video sequence.
  • Finally, it is the most lightweight setup.

The disadvantages are the following ones:

  • While recording, you have a live view of the camera images, however, no preview of what the 3D scan or the depth will look like.
  • With an Nvidia-enable recording PC, H.264 or H.265 compressions can be used to lower the SVO file size, however, without Nvidia capabilities, the SVO file size will be larger due to less-efficient compression.

For example, we recommend the offline setup for scanning applications.
As a reminder, the ZED SDK can be used on Windows and Ubuntu. (not OSX)