AWS solutions documentation change
Summary
Updated solution overview with technical implementation details, revised benefits section to focus on cost optimization and automation, modified architecture diagrams and descriptions, added Spot Instance support, clarified workflow steps, and updated resource references.
Security assessment
The changes focus on technical implementation details, cost optimization (Spot Instances), and workflow automation. No security vulnerabilities, patches, or incidents are mentioned. Existing security mechanisms (IAM, S3) are referenced but not modified or newly documented.
Diff
diff --git a/solutions/open-source-3d-reconstruction-toolbox-for-gaussian-splats-on-aws/index.md b/solutions/open-source-3d-reconstruction-toolbox-for-gaussian-splats-on-aws/index.md index ecc3f2a8e..a04afabce 100644 --- a//solutions/open-source-3d-reconstruction-toolbox-for-gaussian-splats-on-aws/index.md +++ b//solutions/open-source-3d-reconstruction-toolbox-for-gaussian-splats-on-aws/index.md @@ -7 +7 @@ -This Guidance demonstrates how to create high-quality 3D content from real-world objects and environments. It shows how to address common challenges, such as the need for specialized 3D modeling expertise, expensive equipment, and time-consuming manual processes for creating photorealistic 3D assets. The Guidance helps streamline the complex technical infrastructure required for 3D reconstruction and illustrates methods to improve scalability for processing large volumes of 3D assets. It also shows how to enhance the real-time performance of reconstructed 3D models in various applications. By following this Guidance, you can leverage the benefits of 3D content in areas like e-commerce, digital twins, virtual production, and immersive experiences, without being hindered by traditional obstacles. +This Guidance demonstrates how to create high-quality 3D content from real-world objects and environments without requiring specialized 3D modeling expertise, expensive equipment, or time-consuming manual processes for photorealistic asset creation. Users upload input media to Amazon S3, which triggers an automated workflow orchestrated by AWS Step Functions that handles the entire 3D reconstruction pipeline. The process runs on GPU-based compute nodes using either Amazon SageMaker Training Jobs or AWS Batch Jobs with Spot Instances for cost efficiency, automatically selecting the appropriate instance type based on job requirements and notifying users via email when reconstruction completes. You can streamline 3D asset production for e-commerce, digital twins, virtual production, and immersive experiences while scaling to process large volumes of content with automated workflows that eliminate traditional technical and cost barriers. @@ -11 +11 @@ This Guidance demonstrates how to create high-quality 3D content from real-world -Enable real-time digital experiences +Reduce GPU compute costs @@ -13 +13 @@ Enable real-time digital experiences -Create photorealistic 3D models using Gaussian splats. Power responsive e-commerce displays, digital twins, and virtual production applications. +Choose between on-demand and spot instance compute paths for your 3D reconstruction jobs. Pay only for GPU resources during active training while serverless components eliminate idle infrastructure costs. @@ -15 +15 @@ Create photorealistic 3D models using Gaussian splats. Power responsive e-commer -Reduce 3D production barriers +Automate your 3D reconstruction pipeline @@ -17 +17 @@ Reduce 3D production barriers -Turn entire product catalogs or real-world environments into interactive 3D experiences. Eliminate the need for specialized equipment and technical expertise. +Submit media files and a JSON configuration to trigger a fully orchestrated workflow from input validation through Gaussian Splat generation. Receive email notifications when your 3D assets are ready. @@ -19 +19 @@ Turn entire product catalogs or real-world environments into interactive 3D expe -Scale digital asset creation effectively +Deploy with flexible infrastructure tooling @@ -21 +21 @@ Scale digital asset creation effectively -Process thousands of 3D reconstructions simultaneously with event-driven, auto-scaling GPU infrastructure. Build a modular pipeline that automatically handles multiple steps from media processing to model generation. +Launch the complete event-driven architecture using AWS CDK or Terraform. Focus on generating 3D content while automated container builds and model deployment handle operational setup. @@ -27 +27 @@ Event-driven serverless architecture -This architecture diagram shows an automated deployment of event-driven serverless architecture for user accounts. +This Reference architecture shows automated deployment of guidance event-driven, serverless architecture to user accounts @@ -29 +29 @@ This architecture diagram shows an automated deployment of event-driven serverle -[Download the architecture diagram ](https://d1.awsstatic.com/onedam/marketing-channels/website/aws/en_US/solutions/approved/documents/architecture-diagrams/open-source-3d-reconstruction-toolbox-for-gaussian-splats-on-aws.pdf)  Step 1 +[Download the architecture diagram ](downloads/open-source-3d-reconstruction-toolbox-for-gaussian-splats-on-aws.pdf)  Step 1 @@ -31 +31 @@ This architecture diagram shows an automated deployment of event-driven serverle -An administrator deploys the Guidance to an AWS account and Region using AWS Cloud Development Kit (AWS CDK) or Terraform. +Administrator user deploys guidance to AWS account and Region using AWS Cloud Development Kit (CDK) or Terraform. @@ -35 +35 @@ Step 2 -The Base AWS CloudFormation stack to deploy will create all the AWS resources needed to host the Guidance. This includes: an Amazon Simple Storage Service (Amazon S3) bucket, AWS Lambda functions, an Amazon DynamoDB table, necessary AWS Identity and Access Management (IAM) permissions, and an Amazon Elastic Container Registry (Amazon ECR) image registry. Additionally, it includes an AWS Step Functions state machine resource ID in Parameter Store, a capability of AWS Systems Manager, and it creates an Amazon Simple Notification Service (SNS) topic. +The Base AWS Cloud Formation stack to deploy will create all of the AWS resources needed to host the guidance. This includes: Amazon Simple Storage Service (S3) bucket, AWS Lambda functions, an Amazon DynamoDB table, necessary AWS Identity and Access Management (IAM) permissions, Amazon Elastic Container Registry (ECR) image registry, AWS Step Functions State Machine resource ID in AWS Systems Manager Parameter Store, and an Amazon Simple Notification Service (SNS) topic created. @@ -39 +39 @@ Step 3 -Once the Base CloudFormation stack has been deployed, deploy the Post Deploy CloudFormation stack. That stack will build a Docker container and push it to the Amazon ECR registry. It will also build and push the pre-processing models used during training, such as for background removal, into the S3 bucket. +Once the Base AWS Cloud Formation stack has been deployed, the Post Deploy AWS Cloud Formation stack should be deployed. That stack will build a Docker container either locally or using AWS CodeBuild, push it to the Amazon ECR registry, and build and push the preprocessing models used during training (such as for background removal) into Amazon S3 bucket using AWS Lambda. @@ -41 +41 @@ Once the Base CloudFormation stack has been deployed, deploy the Post Deploy Clo -AI-generated open source 3D content +GPU Accelerated Gaussian Splat Reconstruction @@ -43 +43 @@ AI-generated open source 3D content -This architecture diagram shows how to enable the generation of realistic 3D content through open source rendering techniques. +This event-driven, serverless Reference architecture enables the generation of realistic 3D content through cutting-edge, open-source rendering techniques [Download the architecture diagram ](downloads/open-source-3d-reconstruction-toolbox-for-gaussian-splats-on-aws.pdf) @@ -45 +45 @@ This architecture diagram shows how to enable the generation of realistic 3D con -[Download the architecture diagram ](https://d1.awsstatic.com/onedam/marketing-channels/website/aws/en_US/solutions/approved/documents/architecture-diagrams/open-source-3d-reconstruction-toolbox-for-gaussian-splats-on-aws.pdf)  Step 1 + Step 1 @@ -47 +47 @@ This architecture diagram shows how to enable the generation of realistic 3D con -The user authenticates to IAM using AWS Tools and SDKs. +User authenticates to AWS Identity and Access Management (IAM) via AWS Tools and SDKs. @@ -51 +51 @@ Step 2 -The input is uploaded to a dedicated S3 job bucket location. This can be done using a Gradio interface and AWS Software Development Kit (AWS SDK). +The input is uploaded to a dedicated Amazon Simple Storage Service (S3) job bucket location. This can be done using a Gradio interface and AWS Software Development Kit (SDK). @@ -55 +55 @@ Step 3 -Optionally, the Guidance supports external job submission by uploading a '.JSON' job configuration file and media files into a designated S3 job bucket location. +Optionally, the solution supports external job submission by uploading a '.JSON' job configuration file and media files into a designated S3 job bucket location. @@ -59 +59 @@ Step 4 -The job JSON file uploaded to the S3 job bucket will trigger an Amazon SNS message that will invoke the initialization Job Trigger Lambda function. +The job JSON file uploaded to the S3 job bucket will trigger an Amazon Simple Notification Service (SNS) message that will invoke the initialization Job trigger AWS Lambda function. @@ -63 +63 @@ Step 5 -The Job Trigger Lambda function will perform input validation and set appropriate variables for the Step Functions State Machine. +The Job trigger AWS Lambda function will perform input validation and set appropriate variables for the AWS Step Function State Machine. @@ -67 +67 @@ Step 6 -The workflow job record will be created in the DynamoDB job table. +The workflow job record will be created in Amazon DynamoDB job table. @@ -71 +71 @@ Step 7 -The Job Trigger Lambda function will invoke Step Functions State Machine to handle the entire workflow job. +The Job trigger AWS Lambda function will invoke an AWS Step Functions State Machine to handle the entire workflow job. @@ -75 +75 @@ Step 8 -An Amazon SageMaker AI Training Job will be submitted synchronously using the state machine built-in wait until completion mechanism. +Based on the JSON, the process will use either an Amazon SageMaker Training Job (On-Demand Instance) or AWS Batch Job (Spot Instance) that will be submitted synchronously using the state machine built-in wait until completion mechanism. @@ -79 +79 @@ Step 9 -The Amazon ECR container image and S3 job bucket model artifacts will be used to deploy a new container on a graphics processing unit (GPU) based compute node. The compute node instance type is determined by the job JSON configuration. +The Amazon Elastic Container Registry (ECR) container image and S3 job bucket model artifacts will be used to deploy a new container on a Graphics Processing Unit (GPU) based compute node with an instance type determined by the job JSON configuration. @@ -83 +83 @@ Step 10 -The container will run the entire pipeline as an Amazon SageMaker AI training job on a GPU compute node. +The container will run the entire pipeline as an Amazon SageMaker training job or AWS Batch job on a GPU compute node. @@ -87 +87 @@ Step 11 -The Job Completion Lambda function will complete the workflow job by updating the job metadata in DynamoDB and using Amazon SNS to notify the user through email upon completion. +The Job completion AWS Lambda function will complete the workflow job by updating the job metadata in DynamoDB and notifying the user via email upon completion using Amazon SNS. @@ -91 +91 @@ Step 12 -The internal workflow parameters are stored in Parameter Store during deployment to decouple the Job Trigger Lambda function and the Step Function State Machine. +Internal workflow parameters are stored in AWS System Manager Parameter Store during guidance deployment to decouple the Job trigger AWS Lambda function and the AWS Step Function State Machine. @@ -95 +95 @@ Step 13 -Amazon CloudWatch logs and monitors the training jobs, surfacing possible errors to the user. +Amazon CloudWatch is used to log and monitor the training jobs, surfacing possible errors to the user.