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

To learn distinguishable patterns, most of recent works in vehiclere-identification (ReID) struggled to redevelop official benchmarks to providevarious supervisions, which requires prohibitive human labors. In this paper,we seek to achieve the similar goal but do not involve more human efforts. Tothis end, we introduce a novel framework, which successfully encodes bothgeometric local features and global representations to distinguish vehicleinstances, optimized only by the supervision from official ID labels.Specifically, given our insight that objects in ReID share similar geometriccharacteristics, we propose to borrow self-supervised representation learningto facilitate geometric features discovery. To condense these features, weintroduce an interpretable attention module, with the core of local maximaaggregation instead of fully automatic learning, whose mechanism is completelyunderstandable and whose response map is physically reasonable. To the best ofour knowledge, we are the first that perform self-supervised learning todiscover geometric features. We conduct comprehensive experiments on three mostpopular datasets for vehicle ReID, i.e., VeRi-776, CityFlow-ReID, andVehicleID. We report our state-of-the-art (SOTA) performances and promisingvisualization results. We also show the excellent scalability of our approachon other ReID related tasks, i.e., person ReID and multi-target multi-camera(MTMC) vehicle tracking. The code is available at https://github.com/ming1993li/Self-supervised-Geometric.

Self-supervised Geometric Features Discovery via Interpretable Attentio for Vehicle Re-Identification and Beyond (Complete Version)