Active View Planning By Deep Learning
AssigneeRicoh Company, Ltd.
InventorsIvana Tosic, Logan Courtney, Noah Bedard
Life of this patent
Estimating the term…
Abstract
The disclosure includes a system and method for identifying an object and a viewpoint from an image with a probability that satisfies a predefined criterion based on deep network learning. An active view planning application receives a first image, performs recognition on the first image to determine an object, a viewpoint and a probability of recognition, determines a first expected gain in the probability of recognition when a first action is taken and a second expected gain in the probability of recognition when a second action is taken, and identifies a next action from the first action and the second action based on an increase in expected gains.
Claims20
A computer-implemented method comprising:
- receiving, by a computing device, a first image;
- performing, by the computing device, recognition with a deep neural network on the first image to determine an object, a viewpoint and a probability of recognition;
- determining, by the computing device, a first expected gain in the probability of recognition when a first action is taken and a second expected gain in the probability of recognition when a second action is taken;
- and
- identifying a next action from the first action and the second action based on an increase in expected gains.
A system comprising:
- one or more processors;
- and
- a memory, the memory storing instructions, which when executed cause the one or more processors to:
- receive a first image;
- perform recognition using a deep neural network on the first image to determine an object, a viewpoint and a probability of recognition;
- determine a first expected gain in the probability of recognition when a first action is taken and a second expected gain in the probability of recognition when a second action is taken;
- and
- identify a next action from the first action and the second action based on an increase in expected gains.
A computer program product comprising a non-transitory computer readable medium storing a computer readable program, wherein the computer readable program when executed causes a computer to:
- receive a first image;
- perform recognition using a deep neural network on the first image using to determine an object, a viewpoint and a probability of recognition;
- determine a first expected gain in the probability of recognition when a first action is taken and a second expected gain in the probability of recognition when a second action is taken;
- and
- identify a next action from the first action and the second action based on an increase in expected gains.
Dependencies are inferred from the claim text.