Nvidia corporation (20240317263). VIEWPOINT-ADAPTIVE PERCEPTION FOR AUTONOMOUS MACHINES AND APPLICATIONS USING REAL AND SIMULATED SENSOR DATA simplified abstract
VIEWPOINT-ADAPTIVE PERCEPTION FOR AUTONOMOUS MACHINES AND APPLICATIONS USING REAL AND SIMULATED SENSOR DATA
Organization Name
Inventor(s)
Tae Eun Choe of Belmont CA (US)
Minwoo Park of Saratoga CA (US)
Jung Seock Joo of Los Altos CA (US)
VIEWPOINT-ADAPTIVE PERCEPTION FOR AUTONOMOUS MACHINES AND APPLICATIONS USING REAL AND SIMULATED SENSOR DATA - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240317263 titled 'VIEWPOINT-ADAPTIVE PERCEPTION FOR AUTONOMOUS MACHINES AND APPLICATIONS USING REAL AND SIMULATED SENSOR DATA
Simplified Explanation: The patent application discusses systems and methods for adapting perception in autonomous machines using a 3D perception network trained with real and simulated data.
- The 3D perception network can handle unavailable target rig data by training on a combination of real and simulated data.
- Feature statistics from real data are used to transform features from simulated data during training.
- The network is trained alternately on real and simulated data to update shared weights for different paths.
- The resulting network designates one or more paths as the 3D perception network, which can then perform perception tasks on target rig data.
Key Features and Innovation:
- Adapted perception for autonomous machines using a 3D perception network.
- Training network with real and simulated data to handle unavailable target rig data.
- Feature statistics used to transform features during training.
- Alternating training on real and simulated data to update shared weights.
- Designating paths for perception tasks on target rig data.
Potential Applications: The technology can be applied in autonomous vehicles, robotics, surveillance systems, and industrial automation for enhanced perception capabilities.
Problems Solved: The technology addresses the challenge of handling unavailable target rig data in autonomous machines by training the 3D perception network with a combination of real and simulated data.
Benefits:
- Improved perception capabilities for autonomous machines.
- Enhanced adaptability to handle unavailable target rig data.
- Efficient training methods using real and simulated data.
Commercial Applications: The technology can be utilized in the automotive industry for autonomous driving systems, in manufacturing for robotic applications, and in security systems for surveillance tasks.
Questions about Adapted Perception for Autonomous Machines: 1. How does the 3D perception network handle unavailable target rig data? 2. What are the potential applications of this technology in industrial automation?
Frequently Updated Research: Stay updated on advancements in 3D perception networks and training methods for autonomous machines to ensure optimal performance and adaptability.
Original Abstract Submitted
systems and methods are disclosed relating to viewpoint adapted perception for autonomous machines and applications. a 3d perception network may be adapted to handle unavailable target rig data by training one or more layers of the 3d perception network as part of a training network using real source rig data and simulated source and target rig data. feature statistics extracted from the real source data may be used to transform the features extracted from the simulated data during training. the paths for real and simulated data through the resulting network may be alternately trained on real and simulated data to update shared weights for the different paths. as such, one or more of the paths through the training network(s) may be designated as the 3d perception network, and target rig data may be applied to the 3d perception network to perform one or more perception tasks.
