Back to results
US 11,531,354 B2Granted
Image processing apparatus and image processing method
AssigneeSony Corporation
InventorShingo Tsurumi
Life of this patent
Estimating the term…
Abstract
Provided is a map generation unit that generates an attribute-attached occupancy grid map including an existence probability of an obstacle in a space around a moving body for each grid, and an attribute of the obstacle labelled in the occupancy grid map. A position estimation unit that estimates a position of the moving body by matching in a shape of a non-moving body and the attribute between the attribute-attached occupancy grid map and a pre-map that is the attribute-attached occupancy grid map prepared beforehand.
Claims10
An information processing apparatus, comprising:
- a storage device;
- and
- a central processing unit (CPU) configured to:
- receive an image associated with a moving body from a first sensor;
- receive, from a second sensor, distance information indicating a distance of the moving body to an obstacle in a space around the moving body;
- generate depth data based on the received image and the distance information;
- generate, based on the received image, motion data of the moving body and semantic segmentation data;
- generate a first attribute-attached occupancy grid map;
- control storage of the first attribute-attached occupancy grid map in the storage device,
- wherein the first attribute-attached occupancy grid map includes a first edge portion between the obstacle and a free space that does not include the obstacle;
- generate a second attribute-attached occupancy grid map that includes:
- an existence probability of the obstacle in the space around the moving body for each grid of a plurality of grids;
- and
- an attribute of the obstacle labelled,
- wherein the generation of the second attribute-attached occupancy grid map is based on the depth data, the motion data, and the semantic segmentation data;
- extract, from the generated second attribute-attached occupancy grid map, a second edge portion between the obstacle and the free space that does not include the obstacle;
- generate an edge image that includes the second edge portion;
- and
- estimate a first position of the moving body by matching between the edge image and the first edge portion of the first attribute-attached occupancy grid map.
16 more elements
An information processing method, comprising:
- receiving an image associated with a moving body from a first sensor;
- receiving, from a second sensor, distance information indicating a distance of the moving body to an obstacle in a space around the moving body;
- generating depth data based on the received image and the distance information;
- generating, based on the received image, motion data of the moving body and semantic segmentation data;
- generating a first attribute-attached occupancy grid map;
- storing the first attribute-attached occupancy grid map in a storage device,
- wherein the first attribute-attached occupancy grid map includes a first edge portion between the obstacle and a free space that does not include the obstacle;
- generating a second attribute-attached occupancy grid map that includes:
- an existence probability of the obstacle in the space around the moving body for each grid of a plurality of grids;
- and
- an attribute of the obstacle labelled,
- wherein the generation of the second attribute-attached occupancy grid map is based on the depth data, the motion data, and the semantic segmentation data;
- extracting, from the generated second attribute-attached occupancy grid map, a second edge portion between the obstacle and the free space that does not include the obstacle;
- generating an edge image that includes the second edge portion;
- and
- estimating a position of the moving body by matching between the edge image and the first edge portion of the first attribute-attached occupancy grid map.
13 more elements
A non-transitory computer-readable medium having stored thereon computer-executable instructions which, when executed by a processor, cause the processor to execute operations, the operations comprising:
- receiving an image associated with a moving body from a first sensor;
- receiving, from a second sensor, distance information indicating a distance of the moving body to an obstacle in a space around the moving body;
- generating depth data based on the received image and the distance information;
- generating, based on the received image, motion data of the moving body and semantic segmentation data;
- generating a first attribute-attached occupancy grid map;
- storing the first attribute-attached occupancy grid map in a storage device,
- wherein the first attribute-attached occupancy grid map includes a first edge portion between the obstacle and a free space that does not include the obstacle;
- generating a second attribute-attached occupancy grid map that includes:
- an existence probability of the obstacle in the space around the moving body for each grid of a plurality of grids;
- and
- an attribute of the obstacle labelled,
- wherein the generation of the second attribute-attached occupancy grid map is based on the depth data, the motion data, and the semantic segmentation data;
- extracting, from the generated second attribute-attached occupancy grid map, a second edge portion between the obstacle and the free space that does not include the obstacle;
- generating an edge image that includes the second edge portion;
- and
- estimating a position of the moving body by matching between the edge image and the first edge portion of the first attribute-attached occupancy grid map.
13 more elements
Dependencies are inferred from the claim text.