search

MarekKowalski / LiveScan3D

LiveScan3D is a system designed for real time 3D reconstruction using multiple Azure Kinect or Kinect v2 depth sensors simultaneously at real time speed.
MIT License
749 stars 202 forks source link

adding audio #73

Open hrhouma opened 2 months ago

hrhouma commented 2 months ago

Hello, Could you provide any guidance or suggestions on how best to synchronize video and images with audio tracks, for a precise alignment between visual and auditory elements. Thank you for your support and for developing such a versatile tool. I look forward to any recommendations you might have.

ChristopherRemde commented 2 months ago

Hi!

This is indeed somewhat unsolved at the moment. If you take a volumetric video with the Azure Kinects and Temporal Hardware Sync enabled, you should get a video that runs mostly perfectly on time (no dropped frames). If you do some "analoge" syncing, like clapping your hands, and then adjusting the audio clip manually, you should have pretty nicely synced recordings. Especially for short clips (<3 minutes) this should work reasonably well.

Could this maybe help you?

hrhouma commented 2 months ago

Hi,

Thanks for your insights! Just to confirm, when you mention "clapping hands," you're referring to the method of creating a loud clap sound at the beginning of our recording sessions, correct? This sound serves as a manual synchronization point for aligning the color and depth video tracks with the audio, right?

Looking forward to your confirmation or any further advice you might have.

ChristopherRemde commented 2 months ago

Hey! Yes that's exactly what I meant :)

Official audio support will come eventually, but I wouldn't count on it in the next time!