Vehicle Re Identification
Vehicle Re-Identification is an important task in the field of computer vision, aiming to identify different images of the same vehicle captured by non-overlapping camera networks. The core objective of this task is to utilize advanced image processing and deep learning techniques to achieve cross-camera vehicle identity matching and tracking, which has significant application value for intelligent transportation systems, security surveillance, and urban management and optimization.
CityFlow
A Strong Baseline
VehicleID
VehicleNet
VehicleID Large
Recall@k Surrogate loss (ResNet-50)
VehicleID Medium
Recall@k Surrogate loss (ResNet-50)
VehicleID Small
Recall@k Surrogate loss (ResNet-50)
VeRi
vehiclenet
VeRi-776
MBR4B-LAI (w/ RK)
VeRi-Wild Large
VeRi-Wild Medium
VeRi-Wild Small
MBR-4B (without RK)
VRAI test
VRAI test-dev