Navigation

 ·   Wiki Home
 ·   Wiki Help
 ·   Categories
 ·   Title List
 ·   Uncategorized Pages
 ·   Random Page
 ·   Recent Changes
 ·   RSS
 ·   Atom
 ·   What Links Here

Search:

 

Create or Find Page:

 

View CV-VRCam

Overview

CV-VRCam makes it easy to render Cinema 4D scenes as 360 degree Virtual Reality video. CV-VRCam works as a render effect Cinema 4D’s Standard and Physical Render Engine, and outputs Equirectangular, Omni-directional Stereo (stereo Equirectangular), or horizontal cubic strip maps.

For general usage, see the following Cineversity tutorials:

Functionality

CV-VRCam Render Effect

Mode

Omni-Directional Stereo / Equirectangular

This mode renders a map in equirectangular format, which can be projected on a sphere for a 360 view. When rendered in stereo, the rays are adjusted to maintain the correct stereo effect to the left, right and back.

Choose this option for most VR video applications, including YouTube VR, Facebook and MilkVR

Cubic / Horizontal Strip

This mode renders a cubic map as a single horizontal strip, with 6 panes for monoscopic content and 12 panes for stereo content. The image is mapped onto a cube by the viewer. This format offers higher resolution and less distortion than equirectangular projection, but is less compatible especially for video applications.

Choose this option primarily for still images to be viewed on GearVR via Oculus 360 Photos or Orbx.

Cubic Mapping and Anti-Aliasing

Note that visible seams may appear when rendering cubic maps with anti-aliasing or Physical Render, because Cinema 4D’s AA will smear pixels between cube faces.

To minimize this, we recommend the following anti-aliasing settings:

  • Geometry mode
  • Filter: Cubic, Sinc, or Triangle
  • Custom Size: Enabled
  • Filter Width: 0.5
  • Clip Negative Component: Enabled
  • MIP Scale: 0%

In many scenes, it’s possible and necessary to use higher AA settings. In our testing, Best AA can be used as long as the filter width does not exceed 0.5 (Sinc) or 1.0 (Cubic). The filter height can be set higher in order to achieve the desired level of anti-aliasing (since the seams are all horizontal, height blurring does not create visible seams).

Another effective solution is to render with minimal AA at twice the resolution (36864x3072) and scale down. This is the only option to eliminate visible seams when using Physical Render.

Render Stereo

None

Renders a mono 360 image

Both

Renders a stereo 360 image

Left Eye / Right Eye

Renders a mono 360 image based on a single eye and the specified eye separation. Can be used when combining renders later in a compositing pipeline

Eye Separation

The interpupilary distance between the eyes for stereo effect. Average human IPD is 6.5 cm. This can be adjusted for desired audience (smaller for children), or to compensate for the scale of the scene.

Eye to Neck Distance

This is the distance from the eyes to center of the neck, used in the cubic calculation

Cubic Edge Fixup

This option offers some adjustments to the calculation of cubic maps in order to minimize seam artifacts. In general, the Warp option is considered best, but the other options are provided for your experimentation in case they work better on specific scenes.

Pole Merge (Top / Bottom)

Individual options for pole merging at the top and bottom of the image. Pole merging reduces the stereo offset of the cameras towards the poles in order to minimize distortion.

Merge Mode

None

No pole merging is performed. The stereo offset is used for the entire image.

Linear

The stereo offset is reduced in a linear fashion.

Exponential

The stereo offset is reduced exponentially, creating a smoother transition.

Merge Angle

The angle at which the pole merging effect should begin.

Merge Exponent

When using Exponential Merge mode, the exponent controls the falloff ratio of the pole merging effect.

Create Render Settings

This command adds preset render settings in the current C4D document which provide common output sizes and CV-VRCam options for popular VR formats.

Merge Camera Rig

This command merges a camera rig with cameras facing each cubic direction (forward, back, left, right, up, down). Each camera has a 90deg FOV. Currently the rig itself is simply for utility while designing your VR environment and is not necessary for rendering.

Team Render / Network Rendering Support

Network rendering is extremely helpful due to the large output size of VR imagery. CV-VRCam fully supports Cinema 4D’s Team Render solution and command-line rendering. Simply install CV-VRCam within the plugins folder on all render nodes.

Limitations

  • CV-VRCam works with the Standard and Physical Render Engine only. Third-party render engines are not supported (though most offer some form of native VR rendering).
  • Sketch & Toon and Hair rendering is not currently supported.
  • Hardware and Software preview renders are not currently supported.

Support

For CV-VRCam support, please post in the Cineversity Toolbox Forums

Release History

1.5

  • Pole-merging to minimize stereo distortions at the poles
  • Output Cubic Maps in Horizontal Strip format (GearVR)
  • Camera orientation is now evaluated
  • Output Left or Right Eye independently

1.1

  • Added compatibility for C4D R15.057 and greater
  • Fixed omni-directional stereo (ODS) calculation
  • Removed C4D Stereo option, due to limitations
  • Fixed minor issues with VRCam rig

1.0

  • Initial Release
  • Monoscopic and Stereoscopic Equirectangular Rendering