XangleCs documentation

Overview

RenderQ is a production companion app for Xangle Camera Server (XangleCS), designed to offload longer-running processing jobs from your live capture environment.

It acts as an intelligent task queue so your team can keep multi-camera operations responsive while automated rendering and post-processing run in parallel.

RenderQ can run on the same machine as XangleCS, but it is best deployed on a separate workstation to protect performance for critical multi-camera workflows.

What RenderQ Does

• Queues and prioritizes processing jobs by rating or by collection.
• Orchestrates both local applications and cloud services from one task list.
• Supports local tools such as RealityScan, Postshot, and Adobe After Effects.
• Connects to cloud-based generation APIs such as Kling.

For Multi-Camera Photobooth Arrays

RenderQ helps automate high-impact deliverables from multi-camera captures:

• Automatically train Gaussian Splat models and produce smooth left/right replay output.
• Enable custom 3D camera movement workflows in Postshot (planned following the Postshot v1.1 release).
• Place participants into full 3D environments via After Effects pipelines (using Postshot-trained 3DGS assets and Irrealix import workflows).
• Produce two-camera replay videos through After Effects automation, including visual effects and time controls.
• Generate bullet-time style outputs from as few as two cameras using Kling-powered workflows.

RenderQ also integrates with the XangleCS Sharing module. Participants can register immediately after capture, receive an initial confirmation, and then receive a second email once the final render is ready.

Delivery upload is fully automated to your cloud account (Bunny or Cloudinary). Files go directly to your infrastructure, so the RenderQ team never handles or views participant media.

For Studio Pipelines

RenderQ supports fast iteration and scalable batch operations:

• Rapidly validate 3DGS datasets against a reference shot.
• Batch-process full dataset folders for faster client turnaround.

Deployment Recommendation

For best reliability, run RenderQ on a dedicated processing machine connected to your XangleCS environment. This keeps capture-critical systems focused on acquisition while RenderQ handles compute-heavy rendering and automation tasks.

FAQ

Gaussian Splat vs Kling?

Gaussian Splat

• Local, non-generative workflow.
• Requires an NVIDIA GPU.
• Usually takes 3 to 30 minutes depending on image resolution and training settings.
• Training stays on your machine and source files are not sent to a cloud AI service.

Kling

• Cloud-based generative AI workflow.
• Can create bullet-time style output from only two cameras. 😵
• Easier to deploy, but results are less predictable than local deterministic pipelines.
• Requires cloud processing, so participant consent and liability policies should be reviewed.
• Uses credits from your own Kling account (including available developer credits): https://klingai.com/global/dev

Why is RealityScan required for Gaussian Splat?

• Camera registration can sometimes be done in Postshot, but RealityScan is generally faster and more reliable.
• RealityScan outputs reusable standard registration files from a single reference calibration.
• If camera positions stay fixed, the same registration can be reused across subsequent captures.
• This enables consistent virtual camera paths and repeatable results across events or productions.
• For best accuracy, enable RealityScan markers in Settings > RealityScan.

Which Postshot license is required?

• Postshot Studio is required.
• RenderQ relies on silent automation through the Postshot CLI, which requires Studio features.
• Pricing and licenses: https://www.jawset.com/shop/pricing/

How do I run RenderQ on a separate computer?

• In RenderQ, open Settings > General.
• Replace the XangleCS localhost address with the IP address of the computer running XangleCS.
• You can find that IP in XangleCS under Settings > Network, or in the Display module within Xangle.

How do I create and reuse a 3D virtual camera path in Postshot?

• A 3D virtual camera path is the simplest way to create custom camera movement for Gaussian Splat playback.
• You can create the path from your reference dataset, or from any dataset generated using the same reference.
• If you run a new registration (new reference), you may need to recreate the camera path.
• In Postshot, create keyframes for your camera motion, then export via File > Export Cameras.
• Postshot exports an .abc file, which you can assign to your source folders in the Source tab.

How do I use RenderQ with XangleCS?

From XangleCS

• For multi-camera workflows, set your reference (calibration) dataset in the Library.
• Choose a clean capture with the subject centered in frame for best registration consistency.
• To send datasets into RenderQ processing, either:

In RenderQ

• Open the Source tab and choose your intake mode:
• In the same Source tab, you can also choose which template is applied to the reference dataset (optional).