Introduced in The Visual Object Tracking VOT2016 Challenge Results2015
VOT2016 is a video dataset for visual object tracking. It contains 60 video clips and 21,646 corresponding ground truth maps with pixel-wise annotation of salient objects.
Source: Video Saliency Detection by 3D Convolutional Neural Networks Image Source: https://www.researchgate.net/profile/Mohamed_Abdelpakey/publication/327850473/figure/fig3/AS:674547829338114@1537836143562/Visual-results-on-VOT2016-data-set-for-four-sequences.png
Unknown
