Image processing apparatus, method for controlling the same, and storage medium
An image processing apparatus includes one or more memories storing instructions, and one or more processors executing the instructions to obtain three-dimensional shape data of at least one object, wherein the three-dimensional shape data is generated using images of the at least one object captured from a plurality of directions by a plurality of image capturing units, extract a plurality of parts of the obtained three-dimensional shape data, assign identifiers to the extracted plurality of parts based on positions of the extracted plurality of parts, and track the extracted plurality of parts based on the identifiers and the positions of the extracted plurality of parts.
1 . An image processing apparatus comprising:
one or more memories storing instructions; and
one or more processors executing the instructions to:
obtain three-dimensional shape data of at least one object, wherein the three-dimensional shape data is generated using images of the at least one object captured from a plurality of directions by a plurality of image capturing units; and
assign identifiers to a plurality of parts of the obtained three-dimensional shape data corresponding to a predetermined height from a bottom surface of a circumscribe cuboid in the three-dimensional shape data based on positions of the plurality of parts.
2 . The image processing apparatus according to claim 1 , wherein the one or more processors extract the plurality of parts by clipping some of the three-dimensional shape data.
3 . The image processing apparatus according to claim 1 , wherein the one or more processors extract the plurality of parts corresponding to feet in the three-dimensional shape data.
4 . The image processing apparatus according to claim 1 , the one or more processors further execute the instructions to;
track the obtained three-dimensional shape data based on the identifiers and the positions of the plurality of parts; and
calculate a representative position of parts, among the plurality of parts, to which the same identifier is assigned based on a result of the tracking, and output the representative position as an object position.
5 . The image processing apparatus according to claim 1 , wherein the one or more processors assign an identifier to each of the plurality of parts based on a distance between the plurality of parts of the three-dimensional shape data.
6 . The image processing apparatus according to claim 1 , wherein the one or more processors determine in which circumscribed cuboid in the three-dimensional shape data each of the plurality of parts is included, and assign the same identifier to parts included in the same circumscribed cuboid among the plurality of parts.
7 . The image processing apparatus according to claim 1 , wherein the one or more processors set a region in which a predetermined identifier is to be assigned, and assign the predetermined identifier to parts included in the region among the plurality of parts.
8 . The image processing apparatus according to claim 1 , wherein, based on a user's instruction, the one or more processors reassign the identifiers to the plurality of parts to which the identifiers have been assigned.
9 . The image processing apparatus according to claim 1 , wherein the one or more processors project the plurality of parts onto a two-dimensional plane corresponding to a floor surface to generate a two-dimensional image, clip the plurality of parts as independent regions on the two-dimensional image, and output respective circumscribed rectangles of the independent regions on the two-dimensional image.
10 . The image processing apparatus according to claim 4 ,
wherein the one or more processors further obtain a virtual viewpoint, and
wherein the one or more processors generate a virtual viewpoint image based on the virtual viewpoint, the three-dimensional shape data, and a result of the tracking.
11 . The image processing apparatus according to claim 4 , wherein, based on the identifiers and the object position at a time immediately preceding an image capturing time of the plurality of parts, the one or more processors assign the identifiers to the plurality of parts.
12 . The image processing apparatus according to claim 11 , wherein, based on the identifiers and the positions of the plurality of parts at the immediately preceding time, the one or more processors assign the identifiers to the plurality of parts.
13 . The image processing apparatus according to claim 12 ,
wherein, among the plurality of parts, the one or more processors assign, to a part at a position overlapping the object position at the immediately preceding time, the identifier corresponding to the object position, and then determine whether a position of a part to which no identifier has yet been assigned overlaps one of the positions of the plurality of parts at the immediately preceding time,
wherein, in a case where the position of the part overlaps one of the positions of the plurality of parts at the immediately preceding time, the one or more processors assign, to the part, the identifier of the corresponding part at the immediately preceding time, and
wherein, in a case where the position of the part does not overlap any of the positions of the plurality of parts at the immediately preceding time, the one or more processors assign, to the part, the identifier of a part closest to the part in a predetermined range among the plurality of parts.
14 . The image processing apparatus according to claim 13 , wherein, in a case where, among the plurality of parts, there is a part at a position overlapping a plurality of the object positions at the immediately preceding time, the one or more processors assign no identifier to the part, and do not update position information about all the plurality of object positions.
15 . A method for controlling an image processing apparatus, the method comprising:
obtaining three-dimensional shape data of at least one object, wherein the three-dimensional shape data is generated using images of the at least one object captured from a plurality of directions by a plurality of image capturing units; and
assigning identifiers to a plurality of parts of the obtained three-dimensional shape data corresponding to a predetermined height from a bottom surface of a circumscribe cuboid in the three-dimensional shape data based on positions of the plurality of parts.
16 . A non-transitory computer-readable storage medium storing a program for causing a computer to perform a method for controlling an image processing apparatus, the method comprising:
obtaining three-dimensional shape data of at least one object, wherein the three-dimensional shape data is generated using images of the at least one object captured from a plurality of directions by a plurality of image capturing units; and
assigning identifiers to a plurality of parts of the obtained three-dimensional shape data corresponding to a predetermined height from a bottom surface of a circumscribe cuboid in the three-dimensional shape data based on positions of the plurality of parts.