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

Tracking of multiple objects is an important application in AI City gearedtowards solving salient problems related to safety and congestion in an urbanenvironment. Frequent occlusion in traffic surveillance has been a majorproblem in this research field. In this challenge, we propose a model-basedvehicle localization method, which builds a kernel at each patch of the 3Ddeformable vehicle model and associates them with constraints in 3D space. Theproposed method utilizes shape fitness evaluation besides color information totrack vehicle objects robustly and efficiently. To build 3D car models in afully unsupervised manner, we also implement evolutionary cameraself-calibration from tracking of walking humans to automatically computecamera parameters. Additionally, the segmented foreground masks which arecrucial to 3D modeling and camera self-calibration are adaptively refined bymultiple-kernel feedback from tracking. For object detection/classification,the state-of-the-art single shot multibox detector (SSD) is adopted to trainand test on the NVIDIA AI City Dataset. To improve the accuracy on categorieswith only few objects, like bus, bicycle and motorcycle, we also employ thepretrained model from YOLO9000 with multi-scale testing. We combine the resultsfrom SSD and YOLO9000 based on ensemble learning. Experiments show that ourproposed tracking system outperforms both state-of-the-art of tracking bysegmentation and tracking by detection.

Multiple-Kernel Based Vehicle Tracking Using 3D Deformable Model and Camera Self-Calibration