Seeing Machines launches Technical Paper demonstrating real-time 3D human understanding for Physical AI

Seeing Machines

Delivering up to 180 FPS on NVIDIA Jetson Thor, Seeing Machines HMR combines state-of-the-art accuracy with low latency and compact model size for embedded deployment

CANBERRA, Australia, Sept. 30, 2026 (GLOBE NEWSWIRE) — Seeing Machines Limited (AIM: SEE), the advanced computer vision company that has helped redefine transport safety through AI-powered human sensing, today published a Technical Paper demonstrating how robots and other Physical AI systems can achieve real-time 3D human understanding directly on embedded hardware, enabling safer and more natural human-machine interaction.

The paper presents Seeing Machines’ Human Mesh Recovery (HMR) technology, a core capability of the Company’s Human-Centred Physical AI Platform, launched in August 2026. Using a single camera view, HMR reconstructs a detailed 3D representation of a person so that a machine can understand their location, posture and movement in real time. This capability is designed for environments where robots and people need to work safely and effectively together, including factories, hospitals, warehouses and other dynamic environments.

Physical AI systems operating around people must do more than detect a person’s presence. To respond safely and effectively as interactions unfold, robots need to understand a person’s position, posture and movement in real time. Until now, high-fidelity 3D human reconstruction has typically required significant computing resources, making real-time deployment on embedded devices challenging.

Seeing Machines’ HMR technology addresses this constraint by delivering detailed 3D human understanding at the edge. Processing information locally allows robots to respond to a person’s position and movement in real time without relying on a remote cloud connection.

The paper demonstrates Seeing Machines HMR’s compelling balance of accuracy, inference speed and model efficiency when deployed on the NVIDIA Jetson Thor hardware. The results demonstrate that high-fidelity 3D human reconstruction can be performed in real time on embedded hardware previously considered unsuitable for this level of scene understanding.

The HMR model supports both standard RGB cameras and RGB-D cameras without increasing its parameter count.

Paul McGlone, CEO of Seeing Machines, said:”For robots to work safely and effectively alongside people, they need to understand more than simply whether someone is present. Our HMR technology delivers a compelling combination of accuracy, latency and model size on embedded hardware, bringing detailed 3D human understanding into real-world environments. This research demonstrates how Seeing Machines’ 25 years of human-sensing expertise can support the next generation of Physical AI.”

HMR is a core capability within Seeing Machines’ broader Human-Centred Physical AI Platform, which combines the company’s expertise in computer vision, human behaviour and safety-critical AI to enable machines to interpret people and their environments in three dimensions.

Seeing Machines brings more than 25 years of experience developing and deploying Human-Centred AI solutions across automotive, commercial fleet, aviation, rail and off-road applications, providing an established foundation for its expansion into robotics and industrial automation.

Additional Resources:

  • Technical Paper can be found here.
  • Video can be found here.

About Seeing Machines (AIM: SEE): Seeing Machines is a global technology company founded in 2000 and headquartered in Australia, pioneering Human-Centred AI that enables machines to understand people and the environments in which they operate. Combining more than 25 years of Human Factors research with advanced AI, the company develops Safety AI technology that helps machines anticipate risk, make safer decisions in real time, and enable more natural, intuitive human-machine interactions. Seeing Machines combines AI algorithms, embedded processing and optics across its Cabin AI, Cockpit AI and Physical AI platforms, serving automotive, commercial transport, aviation, robotics and industrial automation. The company operates across Australia, the USA, Europe and Asia, supplying technology solutions and services to industry leaders worldwide. www.seeingmachines.com

/Public Release.