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

We focus on the problem of detecting traffic events in a surveillancescenario, including the detection of both vehicle actions and trafficcollisions. Existing event detection systems are mostly learning-based and haveachieved convincing performance when a large amount of training data isavailable. However, in real-world scenarios, collecting sufficient labeledtraining data is expensive and sometimes impossible (e.g. for traffic collisiondetection). Moreover, the conventional 2D representation of surveillance viewsis easily affected by occlusions and different camera views in nature. To dealwith the aforementioned problems, in this paper, we propose a training-freemonocular 3D event detection system for traffic surveillance. Our systemfirstly projects the vehicles into the 3D Euclidean space and estimates theirkinematic states. Then we develop multiple simple yet effective ways toidentify the events based on the kinematic patterns, which need no furthertraining. Consequently, our system is robust to the occlusions and theviewpoint changes. Exclusive experiments report the superior result of ourmethod on large-scale real-world surveillance datasets, which validates theeffectiveness of our proposed system.

Training-free Monocular 3D Event Detection System for Traffic Surveillance