Recording animation
Short overview
Recording an animation in our software is a simple process, and everything happens within the Recording tab. All you need to do is fill in a few straightforward fields, such as scene which will hold our recording, recording format, length and load the calibration files into the device(s).
Position yourself in front of the camera(s), ideally in an A-pose, and go! Once the recording is finished, you can review the results in the scene tab, where post-processing and exporting will take place.
After the recording is finished, you can proceed to post-processing.
Below is a detailed explanation of the entire process, along with important notes and warnings.
Enable Your Device
Navigate to the Recording Tab.
Click on the camera icon next to the device you want to record with. The enabled device will turn green.
Take note of the wrench icon located under the camera; clicking this opens the camera settings where you can tweak exposed camera parameters and load intrinsic calibration.
We’ll refer back to these settings often throughout this guide.
Loading calibrations
Single Camera: It’s recommended to load the calibration into the device during recording so the software is able to track distance of the actor from the camera. While this can be side-loaded later during post-processing, caching it during recording automatically selects it for you later.
Multi-Camera: Both intrinsic and extrinsic calibrations must be loaded before recording begins. Keep in mind that extrinsic calibration is tied to the scene setup - if the position or rotation of any device has changed since the last calibration it will need to be recalibrated.
Refer to this page on how to load an existing calibration data into the setup.
When both calibrations are loaded, a two-color border around camera is displayed.
Set framerate and resolution then launch preview of all cameras to confirm the following:
- Lightning conditions: Make sure that the preview is not too dark, the actor should be well lit and clearly visible. Tweak the camera settings to achieve this (wrench icon), exposure and gain are the two settings most related to this.
- The camera(s) are running at the expected frame rate (FPS). Auto-exposure is usually enabled, but we strongly recommend to DISABLE it unless you know exactly how to use it. For more details, see Camera Settings and Lightning on this page.
For multi-camera setups, auto-exposure should always be off, as all cameras must record at a constant frame rate.
To disable auto-exposure simply move one of the associated settings slider and it will turn it self off.
If you can't find the advertised resolution or FPS, go to the camera settings and try changing the 'Image Format' to a different option. Each image format may provide different recording options, so it is recommended to tweak this setting and observe how the available FPS and resolution change. Similarly certain resolutions might only be accessible at specific FPS values. In multi-camera setups only FPS and resolutions available to all cameras are listed.
Setting a scene for recording
A scene stores your recordings and is used to display output and handle post-processing. You can create multiple scenes, like one for each day or a global scene—it’s designed to help manage your recordings.
Select an active scene from the drop-down menu (Recording for scene), which only lists active scenes. If you haven’t created a scene yet, go to File -> New -> Scene in the top menu to create one.
Estimate Right Away: All frames will be processed immediately after recording. If this option is not selected, you can run the estimation during the post-processing stage.
Estimation Mode: This mode allows you to specify what you want the model to estimate. For example, you can choose to focus on estimating just the right hand/left hand, the entire body, or both body and detailed hands (which is the default setting).
Keep Files on Disk: As the name suggests, enabling this option will store your recordings on disk. We use those frames to display a feed in the post-processing window for visualization purposes and give you the flexibility to run estimations again if needed.
Press Start Recording and go! Once capture is finished, you can view your recording in the scene that you've selected previously. It is where post-processing and exporting happens.
Once recording finishes, you can move onto post-processing your animation.
