Image processing method, recording medium, image processing apparatus, and image-capturing apparatus which first detects a region of a person and then detects a state of the person
An image processing method includes acquiring an image; identifying a position of a region of a person in the acquired image; setting an object detection range based on the identified position of the region of the person; and detecting a state of the person based on the set object detection range.
1 . An image processing method comprising:
acquiring an image which is an equirectangular image, the image having a circular view in at least one direction;
identifying a position of a region of a person in the acquired image;
setting an object detection range based on the identified position of the region of the person;
detecting a state of the person based on the set object detection range; and
duplicating one end portion of the image, in at least the object detection range, to obtain a duplicated one end portion to allow the duplicated one end portion to be continuous with another end portion of the image, the duplicating including adding the duplicated one end portion to the another end portion to generate another image,
the method further comprising:
determining position and size of the object detection range based on a size of the region of the person; and
determining whether to perform the duplicating, based on the position and size of the object detection range,
wherein based on determination that the duplicating is to be performed in the determining, the duplicating is performed after the identifying of the position of the region of the person, and
wherein the detecting of the state of the person is performed based on the generated another image and based on the equirectangular image in which the one end portion has been duplicated and the object detection range.
2 . The image processing method according to claim 1 , wherein:
the duplicating is performed before the identifying of the position of the region of the person, and
the identifying of the position of the region of the person is performed based on the generated another image.
3 . The image processing method according to claim 1 , wherein:
the image has an angle of view of 360 degrees at least a direction.
4 . The image processing method according to claim 1 , wherein:
the detecting of the state of the person includes detecting a skeletal frame of the person based on the image and detecting a posture of the person based on the detected skeletal frame.
5 . The image processing method according to claim 1 , wherein the identifying of the position of the region of the person involves one of:
detecting a person based on the image;
detecting a face based on the image; and
obtaining a difference between images of multiple frames continuously captured and detecting a moving object based on the difference.
6 . The image processing method according to claim 1 , further comprising:
controlling the acquiring of the image based on the state of the person.
7 . A non-transitory recording medium storing a computer-readable code for controlling a computer to carry out the method according to claim 1 .
8 . An image processing apparatus comprising:
an image acquisition circuitry configured to acquire an image which is an equirectangular image, the image having a view that circulates in at least one direction;
a position identification circuitry configured to identify a position of a region of a person in the image acquired by the image acquisition circuitry;
a range setting circuitry configured to set an object detection range based on the position of the region of the person, identified by the position identification circuitry;
a detection circuitry configured to detect a state of the person based on the object detection range set by the range setting circuitry;
duplication circuitry configured to duplicate one end portion of the image to obtain a duplicated one end portion, in at least the object detection range, to allow the duplicated one end portion to be continuous with another end portion of the image, the duplication circuitry further configured to add the duplicated one end portion to the another end portion to generate another image;
a size determination circuitry configured to determine position and size of the object detection range based on a size of the region of the person; and
a necessity-or-unnecessity determination circuitry configured to determine whether to perform duplication by the duplication circuitry, based on the position and size of the object detection range,
wherein based on determination of the necessity-or-unnecessity determination circuitry that duplication is to be performed by the duplication circuitry, the duplication circuitry performs the duplication, and
wherein the detection circuitry detects the state of the person based on the generated another image.
9 . The image processing apparatus according to claim 8 ,
wherein the duplication circuitry performs the duplication before the identification of the position identification circuitry, and
wherein the position identification circuitry identifies the position of the region of the person based on said another image generated by the duplication circuitry.
10 . The image processing apparatus according to claim 8 , wherein:
the image has an angle of view of 360 degrees at least a direction.
11 . The image processing apparatus according to claim 8 , wherein;
the detection circuitry detects a skeletal frame of the person based on the image and detects a posture of the person based on the detected skeletal frame.
12 . An image-capturing apparatus comprising:
the image processing apparatus according to claim 8 ; and
an image sensor to capture the image;
wherein the image processing apparatus further includes an image-capturing control circuitry configured to control the image sensor based on the state of the person detected by the image processing apparatus.