TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild
Download

Motivations

A Large-Scale Dataset and Benchmark for Object Tracking in the Wild.

Large Scale Dataset

> 30K Video Sequences

Object Tracking

> 14M Bounding Boxes

In the Wild

Diversity ensured by Youtube

Papers

Please find here details of all our tracking papers.

TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild.

ECCV’18

Context Aware Correlation Filter Tracking

CVPR’17

Target Response Adaptation for Correlation Filter Tracking

ECCV’16

Multi-Template Scale-Adaptive Kernelized Correlation Filters

ICCVW’15

About

Learn more about our story.

  • Dec' 2015

    Paper Accepted in ICCVW’15

    Multi-Template Scale Adaptive Kernelized Correlation Filters paper presented in ICCVW’15

  • Oct' 2016

    Paper accepted in ECCV’16 as Spotlight

    Target Response Adaptation for Correlation Filter Tracking paper presented as spotlight in ECCV’16

  • July 2017

    Paper accepted in CVPR’17 as Oral

    Context-Aware Correlation Filter Tracking paper presented as oral in CVPR’17

  • Sept' 2018

    TrackingNet Release

    An early version of TrackingNet was published and demoed as a tracking benchmarking tool in ECCV’18

  • Be part
    of our
    story!

Our amazing team

We all work at King Abdullah University of Science and Technology (KAUST).

Matthias Mueller

Matthias Mueller

PhD Student

Adel Bibi

Adel Bibi

PhD Student

Silvio Giancola

Silvio Giancola

Post Doctoral Fellow

Salman Alsubaihi

Salman Alsubaihi

Master Student

Bernard Ghanem

Bernard Ghanem

Associate Professor

Useful Links

Contact us

Contact us here for any inquiry.
You can also follow us on SLACK.