AWS Spatial Computing Blog

An image of RapidPipeline Logo

From Local to Global: Scaling 3D Pipelines with AWS and RapidPipeline

At DGG, our mission has always been to help enterprises scale their 3D data workflows, fast, reliably, and at production scale. With the upcoming launch of our new RapidPipeline plugins at SIGGRAPH 2025, we’re opening the door to seamless integration with the tools 3D professionals already use, and setting the stage for scalable, cloud-first workflows […]

Build a Spatial Object Tracking Pipeline with Create ML using Amazon EC2 Mac Instances

Introduction Apple’s Create ML enables developers to train powerful machine learning models without deep machine learning expertise, simplifying the path from concept to production. A new feature with Apple’s Create ML tool enables object tracking to be trained directly from the command line, unlocking faster and more scalable workflows. This is especially valuable for teams […]

An image of an end user's experience of the Amazon VR tour

Build an Immersive Virtual Reality Experience of Amazon Fulfillment Center Tour

Introduction In the rapidly evolving landscape of warehouse efficiency, inventory planning, and supply chain management, Amazon continues to innovate with its Fulfillment Center (FC) model. Amazon offers free in-person and virtual tours at select locations worldwide, to introduce the people, technology, and processes behind customer order preparation. Amazon Tour allows customers to see different types […]

An image of Topia's Virtual Office

Partner Spotlight: Multiplayer Spatial Connection Engine for Non-Games Using a Scalable P2P Service with Topia

Topia’s Virtual Office, Powered by Topia’s Spatial Connection Engine Introduction This blog post explores the possibilities for spatial computing using peer-to-peer (P2P) webRTC and hybrid server-based approaches to state management. Thereby facilitating the transformation of any application into a game-like multiplayer social experience. Multiplayer game servers and their limitations Multiplayer game backends, such as Amazon […]

AI-Powered Construction Document Analysis by Leveraging Computer Vision and Large Language Models

TwinKnowledge collaborated with AWS’s PACE team to develop a scalable computer vision and Large Language Model solution that efficiently processes thousands of architectural drawings while maintaining high accuracy, helping to reduce construction document errors and improve project efficiency through AI-augmented human review processes.

Meet up with Spatial Computing at AWS re:Invent 2024!

Introduction AWS re:Invent is just around the corner, and Spatial Computing specialists will be on hand at the event to share exciting technical presentations, workshops, chalk talks and more. Spatial computing seamlessly blends the digital and physical worlds, allowing us to interact with digital information in three-dimensional space. It encompasses augmented reality (AR), virtual reality […]

AWS and Meta’s GenAI with XR Hackathon: Shaping the Future

AWS and Meta Hackathon organizing team The AWS and Meta teams collaborated to host the first-ever GenAI and XR Hackathon. Held from August 27-29 at Amazon’s East Palo Alto office, this event explored the combined power of AWS’s generative AI (GenAI) technology and Meta’s Presence Platform capabilities. Attendees leveraged AWS GenAI tools, including Amazon Bedrock, […]

The multiple stages of the 3D reconstruction process illustrated using a park bench

3D Gaussian Splatting: Performant 3D Scene Reconstruction at Scale

August 28, 2025: The instructions in this blog post have been implemented in an AWS Solution, Guidance for Open Source 3D Reconstruction Toolbox for Gaussian Splats on AWS, that includes an end-to-end pipeline for 3D reconstruction that is deployable via AWS Cloud Development Kit (CDK) or Terraform. To learn more, refer to the solution’s implementation guide and sample code. […]