Pedestrian re-identification method and device, computer readable storage medium and terminal equipment
The invention belongs to the technical field of pedestrian re-identification, and particularly relates to a pedestrian re-identification method and device, a computer readable storage medium and terminal equipment. According to the method, a loss function used in a model training process is a preset...
Saved in:
Main Authors | , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
26.08.2022
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention belongs to the technical field of pedestrian re-identification, and particularly relates to a pedestrian re-identification method and device, a computer readable storage medium and terminal equipment. According to the method, a loss function used in a model training process is a preset proportion-based triple loss function; the proportion-based triple loss function restrains that the proportion of the positive sample feature distance to the negative sample feature distance is smaller than a preset proportion threshold value; the positive sample feature distance is the distance between the reference image feature and the positive sample image feature; the negative sample feature distance is the distance between the reference image feature and the negative sample image feature. When facing a scene with very small inter-class difference and very large intra-class difference, compared with a triple loss function based on absolute distance in the prior art, the triple loss function based on the propo |
---|---|
Bibliography: | Application Number: CN202210609192 |