Meta Project Aria - Smart Glasses Research Kit
Project Aria by Meta promotes AI and ML research through a specialized kit for partners, featuring glasses, SDK, and cloud services, with collaborations including BMW and Carnegie Mellon University.
Read original article
Project Aria is an initiative by Meta aimed at advancing artificial intelligence (AI) and machine learning (ML) technologies through collaborative research. The Aria Research Kit is available to approved academic and corporate partners, providing them with specialized glasses and a software development kit (SDK) to conduct independent studies. The kit includes tools for data collection, cloud services for machine perception, and a desktop application called Aria Studio for managing recordings and visualizing data. Notable partnerships include collaborations with BMW to explore AR integration in vehicles and Carnegie Mellon University (CMU) to enhance accessibility for individuals with visual impairments through the NavCog project. The Ego4D Consortium, formed by 15 universities, aims to create a comprehensive dataset of egocentric video to improve AI's understanding of human experiences. Researchers interested in machine perception technologies can apply for access to the Aria Research Kit to further their studies.
- Project Aria supports research in AI and ML through a specialized research kit.
- The kit includes glasses, an SDK, cloud services, and a desktop application for data management.
- Partnerships with BMW and CMU focus on practical applications of AR technology.
- The Ego4D Consortium aims to enhance AI's understanding of daily-life activities through egocentric video data.
- Researchers can apply for the Aria Research Kit to explore machine perception technologies.
Related
Show HN: AI assisted image editing with audio instructions
The GitHub repository hosts "AAIELA: AI Assisted Image Editing with Language and Audio," a project enabling image editing via audio commands and AI models. It integrates various technologies for object detection, language processing, and image inpainting. Future plans involve model enhancements and feature integrations.
MIT researchers advance automated interpretability in AI models
MIT researchers developed MAIA, an automated system enhancing AI model interpretability, particularly in vision systems. It generates hypotheses, conducts experiments, and identifies biases, improving understanding and safety in AI applications.
ARIA: An Open Multimodal Native Mixture-of-Experts Model
Aria is an open-source multimodal AI model with 3.9 billion visual and 3.5 billion text parameters, outperforming proprietary models and enhancing capabilities through a four-stage pre-training pipeline.
New Meta FAIR Research and Models
Meta FAIR has released new research artifacts, including Meta Motivo for humanoid agents, Meta Video Seal for video watermarking, and frameworks like Flow Matching and Large Concept Model to enhance AI capabilities.
Armada: Augmented Reality for Robot Manipulation and Robot-Free Data Acquisition
The paper "ARMADA" presents a system using augmented reality for robot teleoperation, enabling high-quality data collection without physical robots. A user study shows live feedback enhances data quality significantly.
- Some commenters question the practicality and efficiency of using AI for localization compared to traditional beacon systems.
- There is disappointment regarding the lack of advanced features, such as a display in the glasses, reminiscent of earlier technologies like Google Glass.
- Concerns about privacy and the potential for misuse of smart glasses are raised, with calls for restrictions in public spaces.
- Commenters express curiosity about the device's specifications and its intended use, particularly regarding data processing capabilities.
- There is a general sentiment that Meta is shifting focus from the metaverse to more practical applications of AI and VR hardware.
16 commentsIn Project Aria video, they claim to have installed beacons at an airport to enable indoor location, only to dismiss it as something that "doesn't scale."
Instead, they say they "trained" an AI model using vision from glasses, allowing for vision-based localization.
So, here’s an honest question: which approach is actually easier, more cost-effective, and energy-efficient?
1) Deploying 100 or even 1,000 wireless, battery-operated beacons that last 5–7 years—something a non-tech person can set up in a day or two.
2) Training an AI model for each airport, then constantly burning compute power from camera-equipped glasses or phones that barely last a few hours.
Thoughts?
[0] https://facebookresearch.github.io/projectaria_tools/docs/te...
It's not a new headset or a protoype for one.
"Egocentric, multi-modal data as available on future augmented reality (AR) devices provides unique challenges and opportunities for machine perception. These future devices will need to be all-day wearable in a socially acceptable form-factor to support always available, context-aware and personalized AI applications. Our team at Meta Reality Labs Research built the Aria device, an egocentric, multi-modal data recording and streaming device with the goal to foster and accelerate research in this area. In this paper, we describe the Aria device hardware including its sensor configuration and the corresponding software tools that enable recording and processing of such data."
https://facebookresearch.github.io/projectaria_tools/docs/te...
Related
Show HN: AI assisted image editing with audio instructions
The GitHub repository hosts "AAIELA: AI Assisted Image Editing with Language and Audio," a project enabling image editing via audio commands and AI models. It integrates various technologies for object detection, language processing, and image inpainting. Future plans involve model enhancements and feature integrations.
MIT researchers advance automated interpretability in AI models
MIT researchers developed MAIA, an automated system enhancing AI model interpretability, particularly in vision systems. It generates hypotheses, conducts experiments, and identifies biases, improving understanding and safety in AI applications.
ARIA: An Open Multimodal Native Mixture-of-Experts Model
Aria is an open-source multimodal AI model with 3.9 billion visual and 3.5 billion text parameters, outperforming proprietary models and enhancing capabilities through a four-stage pre-training pipeline.
New Meta FAIR Research and Models
Meta FAIR has released new research artifacts, including Meta Motivo for humanoid agents, Meta Video Seal for video watermarking, and frameworks like Flow Matching and Large Concept Model to enhance AI capabilities.
Armada: Augmented Reality for Robot Manipulation and Robot-Free Data Acquisition
The paper "ARMADA" presents a system using augmented reality for robot teleoperation, enabling high-quality data collection without physical robots. A user study shows live feedback enhances data quality significantly.