Fast Visual Tracking via Dense Spatio-temporal Context Learning

By Kaihua Zhang, Lei Zhang, Qingshan Liu, David Dapeng Zhang, Ming-Hsuan Yang. European Conference on Computer Vision, 2014

704 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7069a994c150b0228c4e471ca48ed55d7646bc62

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Large Margin Object Tracking with Circulant Feature Maps (2017, background): “Recently, a group of correlation ﬁlter (CF) based trackers [9, 14, 5, 2, 32, 4] have attracted extensive attentions due to their signiﬁcant computational efﬁciency.”
Discriminative Correlation Filter Tracker with Channel and Spatial Reliability (2016, background): “[45] introduced spatio-temporal context learning in the DCFs.”
The Visual Object Tracking VOT2015 Challenge Results (2015, methodology): “Some correlation ﬁlter based trackers maintained a single model for tracking, i.e., KCFv2 (A.2), DSST (A.56), SAMF (A.54), SRDCF (A.30), PTZ-MOSSE (A.12), NSAMF (A.24), RAJSSC (A.34), OACF (A.13), sKCF (A.3), LOFT-Lite (A.37), STC (A.50), MKCF+ (A.27), and several trackers applied multiple templates”
Learning Collaborative Sparse Representation for Grayscale-Thermal Tracking (2016, methodology): “On one hand, thirteen grayscale trackers are presented to demonstrate their performance in challenging scenarios, including DSST [33], RPT [35], MUSTer [34], MEEM [36], PCOM [40], CN [23], STC [37], SCM [16], KCF [24], CT [39], Struck [18], TLD [38] and MIL [19].”
CREST: Convolutional Residual Learning for Visual Tracking (2017, background): “The examples include kernelized correlation filters [17], multiple dimensional features [12, 18], context learning [49], scale estimation [7], re-detection [31], subspace learning [28], shortterm and long-term memory [20], reliable collection [27] and spatial regularization [9].”
LaSOT: A High-quality Large-scale Single Object Tracking Benchmark (2020, background): “, ECO HC [14], DSST [16], CN [20], CSK [38], KCF [39], fDSST [17], SAMF [55], SCT4 [10], STC [90] and Staple [5]) or deep features (e.”
The Visual Object Tracking VOT2016 Challenge Results (2016, methodology): “Submitted by VOT Committee The STC [111] is a correlation filter based tracker, which uses image intensity features.”
Learning Adaptive Discriminative Correlation Filters via Temporal Consistency Preserving Spatial Feature Selection for Robust Visual Object Tracking (2018, background): “For feature representation, contextual feature information was exploited in [39] to achieve spatial-temporal learning.”
Meta-Tracker: Fast and Robust Online Adaptation for Visual Object Trackers (2018, background): “Many hard cases have been tackled by using context information [55, 38], short and long-term memory [35, 23], and scale-estimation [10], just to name a few.”
Multi-cue Correlation Filters for Robust Visual Tracking (2018, neutral): “[54] incorporate context information into filter learning.”
LaSOT: A High-Quality Benchmark for Large-Scale Single Object Tracking (2018, background): “We evaluate 35 algorithms on LaSOT to provide extensive baselines, comprising deep trackers ( e.g ., MDNet [42], TRACA [5], CFNet [50], SiamFC [4], StructSiam [59], DSiam [16], SINT [49] and VITAL [48]), correlation ﬁl-ter trackers with hand-crafted features ( e.g ., ECO HC [7], DSST [8], CN [11], C”
Learning the Model Update for Siamese Trackers (2019, methodology): “s, and environmental aspects such as motion blur or illumination changes. Currently, there are two prevalent tracking paradigms: Siamese tracking methods [1,27,41,50] and tracking-bydetection methods [3,7,11,19,32,47,48]. In this work, we consider Siamese trackers, since they provide competitive acc”
Transferring Rich Feature Hierarchies for Robust Visual Tracking (2015, methodology): “Besides, some methods exploit correlation ﬁlters for the target or context [4, 14, 37].”
Hierarchical Convolutional Features for Visual Tracking (2015, methodology): “of this connection, we learn one adaptive correlation filter [3, 16, 7, 35, 6, 17] over features extracted from each CNN layer and use these multi-level correlation response maps to collaboratively infer the target location.”
Robust Visual Tracking via Convolutional Networks Without Training (2015, neutral): “c in (2) serves as the target template, which should be updated incrementally to accommodate appearance changes over time for robust visual tracking. We use a simple temporal low-pass ﬁltering method [41], c t = (1 ˆ)c t 1 +ˆ^c t 1; (4) where ˆis a learning parameter, c t is the target template at f”
Long-term correlation tracking (2015, methodology): “Tracking results of our LCT algorithm, KCF [11], STC [28], Struck [9] and TLD [14] methods on twelve challenging sequences (from left to right and top to down are Coke, Tiger2, Shaking, Skating1, Trellis, David, Car4, Fleetface, Jumping, Jogging-2, Dog1, and Singer2, respectively).”