| |

Toyota Research Publishes Machine Learning Advances at Vision Conference

TRI Sim2Real 400x275

October 11, 2021

The Toyota Research Institute (TRI) has announced the acceptance of six research papers in the field of machine learning at the International Conference on Computer Vision (ICCV). The research advances understanding across various tasks crucial for robotic perception, including semantic segmentation, 3D object detection, and multi-object tracking.

Over the last six years, TRI said its researchers have made significant strides in robotics, automated driving and materials science, largely due to machine learning – the application of computer algorithms that constantly improve with experience and data.

“Machine learning is the foundation of our research,” said Gill Pratt, CEO of TRI. “We are working to create scientific breakthroughs in the discipline of machine learning itself and then apply those breakthroughs to accelerate discoveries in robotics, automated driving, and battery testing and development.”

As the ICCV started, TRI shared six papers demonstrating its research in machine learning, including geometric deep learning for 3D vision, self-supervised learning and simulation to real or “sim-to-real” transfer.

“Within the field of machine learning, scalable supervision is our focus,” said Adrien Gaidon, head of TRI’s machine learning team. “It is impossible to manually label everything you need at Toyota’s scale, yet this is the state-of-the-art approach, especially for deep learning and computer vision. Thankfully, we can leverage Toyota’s domain expertise in vehicles, robots or batteries to invent alternative forms of scalable supervision, whether via simulation or self-supervised learning from raw data. This approach can boost performance in a wide array of tasks important for automated cars to be safer everywhere anytime, robots to learn faster and battery development to speed up lengthy testing cycles.”

Key findings within the six papers accepted at ICCV include:

  • Geometric self-supervised learning significantly improves sim-to-real transfer for scene understanding. The resulting unsupervised domain adaptation algorithm enables recognizing real-world categories without requiring any expensive manual real-world labels.
  • TRI’s research on multi-object tracking reveals that synthetic data could endow machines with fundamental human cognitive abilities, like object permanence, that are historically hard for machine learning models but second nature for humans. This development increases the robustness of computer vision algorithms, making them more aligned with people’s visual common sense.
  • TRI’s research on pseudo-lidar shows that large-scale self-supervised pre-training considerably boosts performance of image-based 3D object detectors. The proposed geometric pre-training enables training powerful 3D deep learning models from limited 3D labels, which are expensive or sometimes impossible to get from images alone.

A roundup of all of the papers are available here. For more information on the Toyota Research Institute, visit its website here.


Robotics World News

  • Rapid Robotics Partners with Yaskawa Motoman to Expand Solutions for Industrial Robotic Arms

    Rapid Robotics Partners with Yaskawa Motoman to Expand Solutions for Industrial Robotic Arms

    Rapid Robotics announced a partnership with Yaskawa America, Inc., Motoman Robotics division, to bring industrial capabilities to Rapid Robotics’ Rapid Machine Operator (RMO). The company is now integrated with the entire Yaskawa robotics portfolio, and can deliver more speed, higher payload and increased performance to a greater number of North American manufacturers. “A trusted name… Read More…

  • DroneBase Secures $55 Million to Launch Next Stage Advanced Inspection under New Brand Zeitview

    DroneBase Secures $55 Million to Launch Next Stage Advanced Inspection under New Brand Zeitview

    DroneBase has secured $55 million in investment and rebranded as Zeitview. Led by Valor Equity Partners, an investment firm backing technology companies, the round was followed by existing investors Union Square Ventures, Upfront Ventures, Euclidean Capital, Energy Transition Ventures, and Hearst Ventures. The new funding will support the company’s AI-enabled software inspection solutions. Zeitview inspection professionals use rotary wing drones, crewed… Read More…


Products for Robots & Cobots

  • Digi-Key Electronics Adds 550+ New Suppliers and 75K SKUs in 2022

    Digi-Key Electronics Adds 550+ New Suppliers and 75K SKUs in 2022

    Digi-Key Electronics, which offers electronic components and automation products for immediate shipment, added more than 550 new suppliers and 75,000 new SKUs to their core business and to the Fulfilled by Digi-Key program in 2022. Digi-Key is an authorized distributor of electronic components for more than 2,300 industry-leading suppliers, ensuring engineers, designers, and procurement professionals that the… Read More…

  • Rapid Robotics Partners with Yaskawa Motoman to Expand Solutions for Industrial Robotic Arms

    Rapid Robotics Partners with Yaskawa Motoman to Expand Solutions for Industrial Robotic Arms

    Rapid Robotics announced a partnership with Yaskawa America, Inc., Motoman Robotics division, to bring industrial capabilities to Rapid Robotics’ Rapid Machine Operator (RMO). The company is now integrated with the entire Yaskawa robotics portfolio, and can deliver more speed, higher payload and increased performance to a greater number of North American manufacturers. “A trusted name… Read More…