I'm trying to record and play back complete data using the Python interface and an OAK-D-S2 camera. I want to obtain time-synchronized RGB, stereo, and IMU data.
I referred to holistic record and holistic replay examples, but encountered the following problems:
- The highest frame rate is only 17 fps of saved .mp4 files in tar, even when I set a higher fps in requestOutput.
- During replay, I couldn't find a way to make the playback automatically stop after the data finished playing.
- During replay, the frequency of retrieving images from the queue is also less than the frequency of retrieving .mp4 files.
I also found that when I only record data for one image and imu, the frequency is normal, so the problem is most likely due to synchronization.
So I wondering how to achieve my goal? I want to obtain a single frame of image data at a higher frequency (including an RGB value, a left image, and a right image, all sharing the same timestamp).