Back to results
US 11,590,658 B2Granted
Tactile information estimation apparatus, tactile information estimation method, and program
AssigneePreferred Networks, Inc.
InventorsKuniyuki Takahashi, Jethro Eliezer Tanuwijaya Tan
Life of this patent
Estimating the term…
Abstract
According to some embodiments, a tactile information estimation apparatus may include one or more memories and one or more processors. The one or more processors are configured to input at least first visual information of an object acquired by a visual sensor to a model. The model is generated based on visual information and tactile information linked to the visual information. The one or more processors are configured to extract, based on the model, a feature amount relating to tactile information of the object.
Claims34
An apparatus for estimating tactile information, comprising:
- one or more memories;
- and
- one or more processors configured to:
- input at least visual information of a first object into a neural network model, the neural network model having been trained using visual information of a second object and tactile information of the second object, and
- extract and output, based on the neural network model, information relating to tactile information of the first object,
- wherein the visual information of the first object includes at least texture information of a surface of the first object, and the visual information of the second object includes at least texture information of a surface of the second object.
3 more elements
An apparatus for learning a neural network model, comprising:
- one or more memories;
- and
- one or more processors configured to:
- learn the neural network model using visual information of a second object and tactile information of the second object,
- wherein when at least visual information of a first object is inputted to the learned neural network model, the learned neural network model outputs information relating to tactile information of the first object,
- wherein the visual information of the first object includes at least texture information of a surface of the first object, and the visual information of the second object includes at least texture information of a surface of the second object.
3 more elements
A method for estimating tactile information, comprising
- inputting, by one or more processors, at least visual information of a first object into a neural network model, the neural network model having been trained using visual information of a second object and tactile information of the second object, and
- extracting and outputting, by the one or more processors, based on the neural network model, information relating to tactile information of the first object,
- wherein the visual information of the first object includes at least texture information of a surface of the first object, and the visual information of the second object includes at least texture information of a surface of the second object.
A method for learning a neural network model, comprising
- learning, by one or more processors, the neural network model using visual information of a second object and tactile information of the second object,
- wherein when at least visual information of a first object is inputted to the learned neural network model, the learned neural network model outputs information relating to tactile information of the first object,
- wherein the visual information of the first object includes at least texture information of a surface of the first object, and the visual information of the second object includes at least texture information of a surface of the second object.
A system comprising:
- one or more memories;
- one or more visual sensors acquiring visual information of a first object;
- one or more end effectors manipulating the first object;
- and
- one or more processors configured to:
- input at least the visual information of the first object into a neural network model, and extract and output, based on the neural network model, tactile information to control the one or more end effectors, and
- control the one or more end effectors to manipulate the first object based on the extracted information,
- wherein the neural network model has been trained using visual information of a second object and tactile information of the second object,
- wherein the visual information of the first object includes at least texture information of a surface of the first object, and the visual information of the second object includes at least texture information of a surface of the second object.
6 more elements
A method comprising:
- inputting, by one or more processors, at least visual information of a first object into a neural network model, and extracting and outputting, based on the neural network model, tactile information to control one or more end effectors, and
- controlling, by the one or more processors, the one or more end effectors to manipulate the first object based on the extracted information,
- wherein the neural network model has been trained using visual information of a second object and tactile information of the second object,
- wherein the visual information of the first object includes at least texture information of a surface of the first object, and the visual information of the second object includes at least texture information of a surface of the second object.
1 more element
Dependencies are inferred from the claim text.