# www.projectaria.com > AI-optimized mirror of www.projectaria.com containing 40 pages totalling 13,667 words of clean markdown content, structured data, and semantic HTML. Original source: https://www.projectaria.com/. Last updated: 2026-05-01T14:24:30.484Z. Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG). ## Homepage - [Bringing the Human Perspective to AI.](/site-root.html): Project Aria is a research program from Meta, to help build the future responsibly. Project Aria unlocks new possibilities of how we connect with and experience the world. (383 words) ## Articles & Blog Posts - [datasets/apd/index.html](/datasets/apd/index.html) (1 words) - [EgoBlur Model License Agreement](/tools/egoblur/license/index.html): Accelerating AR and AI from a human perspective. (1,626 words) - [research/efm3d/index.html](/research/efm3d/index.html) (1 words) - [cookies/index.html](/cookies/index.html) (1 words) - [tools/index.html](/tools/index.html) (1 words) - [privacy-policy/index.html](/privacy-policy/index.html) (1 words) - [datasets/index.html](/datasets/index.html) (1 words) - [research/index.html](/research/index.html) (1 words) - [challenges/reconstruction/index.html](/challenges/reconstruction/index.html) (1 words) - [news/aria-training-and-evaluation-toolkit-atek/index.html](/news/aria-training-and-evaluation-toolkit-atek/index.html) (1 words) - [Sonata: Advancing Self-Supervised Learning for 3D Point Representations](/news/introducing-sonata/index.html) (1,456 words) - [Aria Everyday Objects (AEO) Dataset License Agreement](/datasets/aeo/license/index.html): A small-scale, real-world Project Aria dataset with high quality 3D OBB annotations (1,278 words) - [Digital Twin Catalog Dataset License Agreement](/datasets/dtc/license/index.html): The world’s highest quality dataset for object reconstruction research (337 words) - [Introducing “Reading Recognition in the Wild”: A dataset for understanding human behaviors during reading from egocentric sensor suite](/news/introducing-reading-recognition-in-the-wild/index.html) (877 words) - [challenges/index.html](/challenges/index.html) (1 words) - [Project Aria Research Community Guidelines](/community-guidelines/index.html): Meta Platforms Inc. provides these Community Guidelines to outline proper use of Aria for data processing and research. (446 words) - [Bridging the Gap: Introducing Boxer for Open-World 2D-to-3D Lifting 🥊](/news/introducing-boxer/index.html): We are excited to announce Boxer: a new, lightweight and modular model designed to lift open-world 2D bounding boxes into high-precision metric 3D oriented bounding boxes (3DBBs), enabling more detailed annotation of objects in 3D space. (582 words) - [EgoBlur: a set of AI models designed to preserve privacy by detecting and blurring PII from images](/tools/egoblur/index.html): EgoBlur is an open source AI model that helps researchers innovate responsibly by detecting and blurring PII from image data. (894 words) - [notices/index.html](/notices/index.html) (1 words) - [ariagen2devicepaper/index.html](/ariagen2devicepaper/index.html) (1 words) - [research/efm3d/index.html](/research/efm3d/index.html) (1 words) - [datasets/aea/index.html](/datasets/aea/index.html) (1 words) - [datasets/ase/index.html](/datasets/ase/index.html) (1 words) - [datasets/adt/index.html](/datasets/adt/index.html) (1 words) - [datasets/nymeria/index.html](/datasets/nymeria/index.html) (1 words) - [datasets/hot3d/index.html](/datasets/hot3d/index.html) (1 words) - [datasets/gen2pilot/index.html](/datasets/gen2pilot/index.html) (1 words) - [resources/index.html](/resources/index.html) (1 words) - [research-kit/index.html](/research-kit/index.html) (1 words) - [Digital Twin Catalog](/datasets/dtc/index.html): The world’s highest quality dataset for object reconstruction research (868 words) - [Where collaboration meets innovation](/partners/index.html): Project Aria Research Kit: Collaboratively supporting the acceleration of AI and ML technology. (570 words) - [Aria Gen 2 applications are open. Our next generation hardware unlocks new possibilities.](/datasets/reading-in-the-wild/index.html) (777 words) - [Photoreal Reconstruction](/photoreal-reconstruction/index.html) (546 words) - [Project Aria accelerates AR and AI research](/our-work/index.html): Project Aria is a research program from Meta, to help build the future responsibly. Project Aria unlocks new possibilities of how we connect with and experience the world. (430 words) - [INTRODUCING ARIA GEN 2 GLASSES](/glasses/index.html): Aria Gen 2 Glasses offer a wearable computer in a glasses form-factor, with on-device compute capabilities for research use cases across robotics and contextual AI. (343 words) - [SceneScript: an AI model and method to understand and describe 3D spaces](/scenescript/index.html): SceneScript is a method for representing and inferring scene geometry using an auto-regressive structured language model, and end-to-end learning. (1,046 words) - [Aria Everyday Objects (AEO)](/datasets/aeo/index.html): A small-scale, real-world Project Aria dataset with high quality static 3D oriented bounding boxes annotation (494 words) - [Thank you for your interest](/contact/index.html): Need to get in touch with us? Get answers to questions about the Project Aria program. (184 words) - [Aria Gen 2 applications are open. Our next generation hardware unlocks new possibilities.](/news/index.html): Project Aria News provides updates on new datasets and challenges. (509 words) ## Resources - [Full Page Index](/index.html): Browse all cached pages with rich metadata - [About This Cache](/content/about.html): Methodology, technical details, and usage guidelines - [XML Sitemap](/content/sitemap.xml): Machine-readable sitemap for crawler discovery - [Robots.txt](/content/robots.txt): Crawler directives