English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In comparison with person re-identification (ReID), which has been widelystudied in the research community, vehicle ReID has received less attention.Vehicle ReID is challenging due to 1) high intra-class variability (caused bythe dependency of shape and appearance on viewpoint), and 2) small inter-classvariability (caused by the similarity in shape and appearance between vehiclesproduced by different manufacturers). To address these challenges, we propose aPose-Aware Multi-Task Re-Identification (PAMTRI) framework. This approachincludes two innovations compared with previous methods. First, it overcomesviewpoint-dependency by explicitly reasoning about vehicle pose and shape viakeypoints, heatmaps and segments from pose estimation. Second, it jointlyclassifies semantic vehicle attributes (colors and types) while performingReID, through multi-task learning with the embedded pose representations. Sincemanually labeling images with detailed pose and attribute information isprohibitive, we create a large-scale highly randomized synthetic dataset withautomatically annotated vehicle attributes for training. Extensive experimentsvalidate the effectiveness of each proposed component, showing that PAMTRIachieves significant improvement over state-of-the-art on two mainstreamvehicle ReID benchmarks: VeRi and CityFlow-ReID. Code and models are availableat https://github.com/NVlabs/PAMTRI.

PAMTRI: Pose-Aware Multi-Task Learning for Vehicle Re-Identification Using Highly Randomized Synthetic Data