Information processing device, information processing method, and program
An object recognition unit detects, from image data that represents an image using respective signal values of a plurality of pixels, a first region that is a region representing the object in the image. The object recognition unit determines a first confidence level that is a class confidence level for a first class, the first class being a class of the object represented in the first region. A region recognition unit segments the image of the image data into second regions representing different classes of an object, and determines, for each of second regions, a second class being a class of an object in the second region. An object determination unit determines, as the second class, a class of an object in a non-overlapping region that is the second regions that do not overlap with the first region.
1. An information processing device comprising:
a memory configured to store instructions; and
a processor configured to execute the instructions to detect an object around the information processing device and determine a class of the object,
wherein the processor is configured to execute the instructions to:
detect, from image data that represents an image using respective signal values of a plurality of pixels, a first region that is a region representing the object in the image, and determine a first confidence level that is a class confidence level for a first class, the first class being a class of the object represented in the first region;
segment the image of the image data into second regions representing different classes of an object, and determine, for each of second region, a second class that is a class of an object in the second region;
determine, as the second class, a class of an object in a non-overlapping region that is the second regions that do not overlap with the first region,
determine, as the second class, a class of an object in an overlapping region that is the second regions that overlap with the first region in case of the first confidence level for the overlapping region lower than a predetermined confidence level threshold; and
determine the class of the object in the overlapping region as the first class in case of the first confidence level for the overlapping region equal to or greater than the predetermined confidence level threshold.
2. The information processing device according to claim 1 , wherein the processor is configured to execute the instructions to:
acquire depth information indicating a depth of an object for each of sample points corresponding to the plurality of pixels; and
adjust the second region so that the depth of the object represented in the second region is maintained within a predetermined range.
3. The information processing device according to claim 2 , wherein the processor is configured to execute the instructions to start the segmentation of the image and start the acquisition of the depth information in case of presence of an object, among detected objects, at a distance from the information processing device being greater than or equal to a predetermined distance threshold.
4. The information processing device according to claim 2 , wherein the processor is configured to execute the instructions to start the segmentation of the image and start the acquisition of the depth information in case of information on the surrounding environment around the information processing device indicating a predetermined environment difficult to detect the first region from the image data.
5. The information processing device according to claim 2 , wherein the processor is configured to execute the instructions to start the segmentation of the image and start the acquisition of the depth information in case of a number of detected objects equal to or greater than a predetermined number threshold or in case of an interval between the first regions of each detected object being equal to or less than a predetermined interval threshold.
6. The information processing device according to claim 2 , wherein the processor is configured to execute the instructions to start the segmentation of the image and start the acquisition of the depth information in case of a class of a detected object being a predetermined class, or in case of the class of the object not determined.
7. The information processing device according to claim 2 , wherein the processor is configured to execute the instructions to start the segmentation of the image and start the acquisition of the depth information in case of a velocity of a vehicle carrying the information processing device becoming equal to or greater than a predetermined velocity threshold, or an estimated time to collide based on a distance from the information processing device to a detected object and the velocity of the vehicle becoming equal to or less than a predetermined threshold for time to collide.
8. An information processing method executed by an information processing device, the information processing method comprising:
detecting, from image data that represents an image using respective signal values of a plurality of pixels, a first region that is a region representing an object in the image, and determining a first confidence level that is a class confidence level for a first class, the first class being a class of the object represented in the first region;
segmenting the image of the image data into second regions representing different classes of an object, and determining, for each second region, a second class being a class of an object in the second region;
determining, as the second class, a class of an object in a non-overlapping region that is the second regions that do not overlap with the first region;
determining, as the second class, the class of the object in an overlapping region that is the second regions that overlap with the first region in case of the first confidence level for the overlapping region lower than a predetermined confidence level threshold; and
determining the class of the object in the overlapping region as the first class in case of the first confidence level for the overlapping region equal to or greater than the predetermined confidence level threshold.
9. A non-transitory computer-readable medium storing a program for causing a computer to execute:
detecting an object around the information processing device and determining a class of the object,
wherein the program causes the computer to execute:
detecting, from image data that represents an image using respective signal values of a plurality of pixels, a first region that is a region representing the object in the image, and determining a first confidence level that is a class confidence level for a first class, the first class being a class of the object represented in the first region;
segmenting the image of the image data into second regions representing different classes of an object, and determining, for each of second region, a second class being a class of an object in the second region;
determining, as the second class, a class of an object in a non-overlapping region that is the second regions that do not overlap with the first region;
determining, as the second class, a class of an object in an overlapping region that is the second regions that overlap with the first region in case of the first confidence level for the overlapping region lower than a predetermined confidence level threshold; and
determining the class of the object in the overlapping region as the first class in case of the first confidence level for the overlapping region equal to or greater than the predetermined confidence level threshold.