Creating & Exporting Animations using VR

How to create movies using the VR Stories module in Zeiss arivis Pro VR

TL;DR

  1. Open your image in arivis Pro VR
  2. Open the Tools menu, then Toolsets, and switch to the VR Stories tool
  3. Start to record you animation by using the Trigger, use the A button to store the animation path
  4. Remove your VR goggles and use the VR Stories panel to review and export your animation as a movie file

Introduction

VR provides a great environment for exploring 3D and 4D data, and sharing this environment using the Collaborative VR module in arivis is the best way to let multiple users enjoy the same benefits of depth perception which makes VR such an asset. However, Since it is hardly practical to equip a complete auditorium with VR headsets at present, creating movies that can be easily added to presentations is probably the next best thing. Since VR provides a great way to navigate through datasets and encourages exploring viewpoints that may me be difficult to conceive using screen based 3D viewers, it is also a great environment to create such animations for export. The VR Stories module in arivis Pro VR provides the tools needed to create animations directly in the VR environment for export as movies.

Recording animations

Notes on recording animations

The VR Stories module provides several options for creating animations. The first of these is whether to record a continuous or step by step recording. In continuous recording the user starts recording and the software automatically and continuously records the user's actions. This method is in some sense easier as there is no need for the user to actively interact with the recording process except to terminate the recording when finished. However, this method functions best if the user has already prepared what they want to show in some detail to avoid having to start over again each time a hesitation or mistake spoils the recording. therefore, some rehearsal is highly recommended.

In contrast, the step by step recording requires that the user actively records each transition step much like the storyboard works in arivis Pro. This is a little more laborious, but gives the user plenty of room to explore and refine each step in the animation to ensure the best results.

VR Stories can include changes in position and scaling of the volume, position of the clipping tool, opacity changes, changes in rendering mode, and object visibility. However, they do not include object segmentation or isosurface modifications, and generally the creation of manual measurements. If we want to record such things it may be best to use screen recording software to capture what is shown in the arivis window, which mirrors what the user sees in the glasses.

The main advantage of creating animations over just recording the screen is the ability the generate higher resolution movies than what we may be capable of displaying in the headset.

As mentioned in our VR introduction page, one key aspect of creating a useful and comfortable VR experience is the ability to render high quality renderings very rapidly through the headset. This is the main reason that arivis VR requires a desktop GPU based rendering. But even with powerful desktop GPUs some compromises must be made to render quality and resolution to guarantee a fast frame rate, more so for large datasets (typically anything larger than 1000x1000x1000 pixel equivalent). But while a high frame rate is clearly important while the users is wearing a headset, once an animation is recorded we can usually afford taking a little more time to generate a movie of higher quality from the animation when the user is no longer wearing the headset. If using screen recording software there is no option to generate higher quality movies than what is visible in the headset. However, exporting animations as movies does provide the option to generate higher resolution renderings.

Recording an animation in VR, especially when in continuous mode, is quite similar to recording footage using a hand-held camera. The continuous recording smoothes out the movement by sampling the position every second, but care should be taken to minimize head movements when recording.

Also, because the VR experience is generally quite fluid, adept users can tend to navigate rather quickly through their datasets, and while this feels very natural with the headset on, it can leads to animations that are much too fast for an outside viewer to follow. Therefore, when recording animations it is recommended to:

  • take your time
  • avoid fast and/or wide movements
  • pause regularly to allow your audience to focus on what you want to show them
  • remember that the controllers are not visible in the exported video so they cannot be used to point at specific structures or objects
  • use object visibility filters from the main menu to allow your audience to focus on specific objects as needed

Accessing the VR Stories tools

The VR Stories tool is accessed through the tools menu in VR.

Creating an animation with continuous recording

To create an animation in continuous mode, start by opening the tools menu. Go to the Recording tools and select the CONT option. Then, use the trigger to start recording. Record your animation then press the A button to finish recording. 

Creating an animation with step-wise recording

Creating an animation in step-wise mode is very similar, in that we start by clicking the trigger button and still use A to terminate the recording, but we must also press the trigger for each additional keyframe.

Example Recording

As mentioned above, an animation in VR can include the following visualisation elements:

  • Volume scale and position
  • Current timepoint (for time series)
  • Clipping planes/volume positions and appearance
  • Rendering mode
  • Opacity settings
  • Object visibility

But does not include:

  • Manual object creation or editing steps
  • Surface transformations
  • User interface elements (menus, controllers, environment)

Bear this in mind when recording an animation. If your movie should include the element not included in animations as mentioned above you should consider screen recording instead. Note that if you use screen recording software, you can switch to full-screen mode by pressing F11 on the keyboard. Pressing F11 again will also switch back out of full-screen.

Here is an example of a video captured using screen recording software:

And here is the animation we created above exported as a movie: 

Exporting animations as movie files

Once the animation recording is finished, we can export it as movie file. this needs to be done outside the VR environment as the GPU will be required to do the processing and will not be available to render the VR view at the same time. After we remove the goggle we can open the VR Stories panel by right-clicking anywhere in the panels on the right of the arivis Window or from the Window> Panels> VR Stories.

In the VR Stories panel we can see a list of the stories created for a specific dataset and we can also change some of the parameters of our recording (e.g. change the name or the duration). Most importantly, we can select any animation recording and export them as a movie file that we can share in presentations, edit using video editing software, export to external media players etc. 

When exporting an animation as video, we have several available options.

Movie Style

We can either export a storyboard that transitions automatically from one keyframe to the next using continuous adjustments, or export a VR Journey of individual snapshots. the example above was produced as a Storyboard, the following example is the same recording exported as a VR journey:

Framerate

A movie file is a series of individual frames which, when played in sequence, produce an animation. The higher the frame rate the smoother the animation will appear, but the file will also be much larger, require a lot more time to produce, and potentially require high-end hardware to playback (double the framerate > double the file size > double the rendering time > double the playback resource usage).

A framerate of 30FPS is usually fine for most uses, but 60FPS may be preferable for 360 degree videos (see below).

A lower framerate will render much faster and may be used to check the animation export before rendering the movie again with higher framerate and resolution settings for production.

Resolution

This dictates the dimension of the output movie file. This affect the sharpness of the video when viewed on a high resolution display, and the rendering time (more pixels to render, more processing time).

For inclusion in a PPT presentation a resolution of 720p is usually enough. 1080p is better if the video will be shown full screen. UHD (4K) and FUHD (8K) are also available and can be useful if you have access to high resolution displays for playback, or if you want to edit the video in a 3rd party video editing software (e.g. splice a VR Story video with other video footage).

In the Resolution option we can also select 360 videos. These videos can be played back in VR style viewers (e.g. google cardboard with a mobile phone), where the person looking at the video can look around in any direction with the view automatically responding to head movements. These videos are clearly not suitable to an auditorium style environment, but can be a great way to showcase your your images. Several free video players support 360 videos, including video sharing websites like YouTube.

Here is the video above rendered as a 360 video.

Note that on a desktop screen you may need to use the mouse cursor to adjust the orientation in the viewer, whereas in 360 video players head movements usually suffice.

Quality

The quality setting is most important for large volumes, ut generally has a big impact on the processing time. The quality setting goes from Very Low to Very High and each increase in quality level results in both an increase in the quality of the video render and the processing time needed to produce the video file.