Tracking More Than 100 Arbitrary Objects at 25 FPS Through Deep Learning

TítuloTracking More Than 100 Arbitrary Objects at 25 FPS Through Deep Learning
AutoresLorenzo Vaquero, Víctor M. Brea, Manuel Mucientes
TipoArtículo de revista
Fonte Pattern Recognition, ELSEVIER SCI LTD, Vol. 121, pp. 108205 , 2022.
RankProvisionally ranked Q1 in Software by CiteScore 2020
DOI10.1016/J.PATCOG.2021.108205
AbstractMost video analytics applications rely on object detectors to localize objects in frames. However, when real-time is a requirement, running the detector at all the frames is usually not possible. This is somewhat circumvented by instantiating visual object trackers between detector calls, but this does not scale with the number of objects. To tackle this problem, we present SiamMT, a new deep learning multiple visual object tracking solution that applies single-object tracking principles to multiple arbitrary objects in real-time. To achieve this, SiamMT reuses feature computations, implements a novel crop-and-resize operator, and defines a new and efficient pairwise similarity operator. SiamMT naturally scales up to several dozens of targets, reaching 25 fps with 122 simultaneous objects for VGA videos, or up to 100 simultaneous objects in HD720 video. SiamMT has been validated on five large real-time benchmarks, achieving leading performance against current state-of-the-art trackers.
Palabras chavemultiple visual object tracking, motion estimation, deep learning, Siamese networks