US 12,367,675 B2
Multi-camera person association via pair-wise matching in continuous frames for immersive video
Longwei Fang, Beijing (CN); Qiang Li, Beijing (CN); Wenlong Li, Beijing (CN); Yikai Fang, Beijing (CN); and Hang Zheng, Beijing (CN)
Assigned to Intel Corporation, Santa Clara, CA (US)
Appl. No. 17/923,883
Filed by Intel Corporation, Santa Clara, CA (US)
PCT Filed Jul. 30, 2020, PCT No. PCT/CN2020/105816
§ 371(c)(1), (2) Date Nov. 7, 2022,
PCT Pub. No. WO2022/021217, PCT Pub. Date Feb. 3, 2022.
Prior Publication US 2023/0237801 A1, Jul. 27, 2023
Int. Cl. G06T 7/292 (2017.01); G06V 10/26 (2022.01); G06V 20/40 (2022.01)
CPC G06V 20/42 (2022.01) [G06T 7/292 (2017.01); G06V 10/267 (2022.01); G06V 2201/07 (2022.01); G06V 2201/12 (2022.01)] 21 Claims
OG exemplary drawing
 
1. A system comprising:
a memory to store bounding box data for a plurality of video frames, ones of the video frames corresponding to respective views of a scene at a first time instance;
instructions; and
at least one processor circuit to be programmed based on the instructions to:
generate a distance matrix based on the bounding box data, the distance matrix including distances between available combinations of bounding boxes between pairs of the video frames;
determine a plurality of distance sub-matrices from the distance matrix based on two-way minimum distance correspondence bounding box pairs within the distance matrix;
determine a plurality of correspondence sub-matrices corresponding respectively to the plurality of distance sub-matrices based on an objective function, the objective function including the plurality of correspondence sub-matrices and the plurality of distance sub-matrices; and
provide an association list for the bounding box data of the video frames, the association list based on the correspondence sub-matrices.