IP Library Granted Patent US 12694644
Granted Patent B2
US 12694644 · App. 18/608,944 · Granted Jul 28, 2026

Re-identification system

Inventors: Quan Kong (Tokyo-to, JP); Hsuan-Kung Yang (Tokyo-to, JP); Norimasa Kobori (Tokyo-to, JP); Jira Jindalertudomdee (Tokyo-to, JP)
Assignee: TOYOTA JIDOSHA KABUSHIKI KAISHA
G06V10/761
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12694644
App. No.
18/608,944
Granted
Jul 28, 2026
Kind
B2
Abstract

A re-identification system temporarily performs a re-identification process for determining whether two moving objects shown in a plurality of videos are identical or not. Similarities between the two moving objects in the re-identification process are ranked in consideration of a direction of each moving object. The rank is highest when the two moving objects are identical and the two moving objects are same in the direction. The rank is lowest when the two moving objects are not identical and the two moving objects are different in the direction. A ranking rule is that the rank is higher as the similarity is higher. The re-identification system calculates a degree of consistency between the ranking result and the ranking rule. Then, the re-identification system finally determines whether the two moving objects are identical or not based on the degree of consistency in addition to the similarities.

Claims (31)

1 . A re-identification system for re-identification of a moving object, the re-identification system comprising:

processing circuitry; and

one or more memories configured to store feature information indicating a combination of a re-identification feature amount and a direction of the moving object in a video, wherein

types of a rank regarding a similarity between two moving objects include:

a highest rank in a case where the two moving objects are identical and the two moving objects are same in the direction;

a lowest rank in a case where the two moving objects are not identical and the two moving objects are different in the direction; and

an intermediate rank between the highest rank and the lowest rank,

a ranking rule is that the rank is higher as the similarity is higher, and

the processing circuitry is configured to:

calculate a plurality of similarities between a first moving object and a second moving object based on a plurality of combinations of the re-identification feature amount of the first moving object and the re-identification feature amount of the second moving object;

perform a first determination process that temporarily determines whether the first moving object and the second moving object are identical or not based on each of the plurality of similarities;

determine a plurality of ranks for the plurality of similarities based on a result of the first determination process, the direction of the first moving object, and the direction of the second moving object;

compare a correspondence relationship between the plurality of similarities and the plurality of ranks with the ranking rule to calculate a degree of consistency between the correspondence relationship and the ranking rule; and

perform a second determination process that determines whether the first moving object and the second moving object are identical or not based on the degree of consistency and the plurality of similarities,

a first weight coefficient is a weight coefficient in a case where the direction of the first moving object and the direction of the second moving object are same,

a second weight coefficient is a weight coefficient in a case where the direction of the first moving object and the direction of the second moving object are different from each other, and the second weight coefficient is smaller than the first weight coefficient, and

the processing circuitry is further configured to:

select, for each of the plurality of similarities, the first weight coefficient or the second weight coefficient as the weight coefficient based on the direction of the first moving object and the direction of the second moving object; and

calculate a score by multiplying each similarity by the weight coefficient.

2 . The re-identification system according to claim 1 , wherein

the processing circuitry is further configured to:

calculate, for each of the plurality of similarities, the score that becomes higher as the similarity becomes higher;

perform the first determination process that temporarily determines whether the first moving object and the second moving object are identical or not based on the score;

calculate a corrected score by correcting the score based on the degree of consistency; and

perform the second determination process that determines whether the first moving object and the second moving object are identical or not based on the corrected score.

3 . The re-identification system according to claim 2 , wherein

the processing circuitry is further configured to:

compare the correspondence relationship between the plurality of similarities and the plurality of ranks with the ranking rule to calculate the degree of consistency for each of the plurality of similarities; and

correct the score by multiplying the score for the similarity by the degree of consistency for each similarity to calculate the corrected score.

4 . The re-identification system according to claim 1 , wherein

the feature information indicates the combination of the re-identification feature amount and the direction of the moving object for a track, the track being information representing a time series of the moving object in the video.