US 11,699,290 B1Granted

Pedestrian re-identification method and apparatus based on local feature attention

AssigneeZhejiang Lab

InventorsHongsheng Wang, Guang Chen

Life of this patent

Estimating the term…

Loading patent…

Abstract

Disclosed are a pedestrian re-identification method and apparatus based on local feature attention. The method includes the following steps: S1: obtaining an original surveillance video image data set, and dividing the original surveillance video image data set into a training set and a test set in proportion; and S2: performing image enhancement on the original surveillance video image training set to obtain enhanced images, and converting the enhanced images into sequence data. The pedestrian re-identification technology based on local feature attention uses a multi-head attention mechanism neural network to capture, extract video image feature sequences and replace convolution kernels in a convolutional neural network, uses fully connected layers and an activation function to combine local pedestrian feature sequences into complete pedestrian feature sequences through a weight matrix, performs prediction on the obtained pedestrian feature sequences, outputs position coordinates of pedestrians in the images and selects pedestrians to realize pedestrian re-identification.

Claims10

Dependent claims

A pedestrian re-identification method based on local feature attention, comprising the following steps:

S1:
obtaining an original surveillance video image data set, and dividing the original surveillance video image data set into a training set and a test set in proportion;
S2:
performing image enhancement on the original surveillance video image training set to obtain enhanced images, and converting the enhanced images into sequence data;
S3:
performing feature sequence extraction on the enhanced image by using a pedestrian re-identification neural network to obtain feature sequences of all objects in the images;
S4:
capturing, from the feature sequences of all objects in the images, local pedestrian feature sequences by using a multi-head attention mechanism;
S5:
combining the local pedestrian feature sequences into complete pedestrian feature sequences through a weight matrix by using fully connected layers and an activation function;
S6:
performing prediction on the obtained pedestrian feature sequences to determine position coordinates of pedestrians in the images;
S7:
calculating a predicted intersection ratio of the pedestrian feature sequences according to the obtained position coordinates of pedestrians and the coordinates of image annotation samples in the original surveillance video image data set;
S8:
iteratively training a pedestrian re-identification neural network by means of residual connection to obtain a pedestrian re-identification detection model;
and
S9:
inputting the test set into the pedestrian re-identification detection model to output pedestrian identification results.

16 more elements

Dependencies are inferred from the claim text.

Facts

Application: 17/954,129Filed 27 Sep 2022 · continuation
Priority: CN 202210714468.123 Jun 2022
Granted: 11 Jul 2023
Term ends: Not estimatedThe source records don’t support an estimate.

Classification21 codes

G06V 20/53
Surveillance or monitoring of activities, e.g. for recognising suspicious objects: recognition of crowd images, e.g. recognition of crowd congestion
G06V 40/103
Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands: static body considered as a whole, e.g. static pedestrian…
G06T 5/92
Dynamic range modification of images or parts thereof: based on global image properties
G06T 7/73
Determining position or orientation of objects or cameras: using feature-based methods