Learning Background-Aware Correlation Filters for Visual Tracking

By Hamed Kiani Galoogahi, Ashton Fagg, Simon Lucey. IEEE International Conference on Computer Vision, 2017

1,139 citations recorded by the source corpus.

Correlation Filters (CFs) have recently demonstrated excellent performance in terms of rapidly tracking objects under challenging photometric and geometric variations. The strength of the approach comes from its ability to efficiently learn - on the fly - how the object is changing over time. A fundamental drawback to CFs, however, is that the background of the target is not modeled over time which can result in suboptimal performance. Recent tracking algorithms have suggested to resolve this drawback by either learning CFs from more discriminative deep features (e.g. DeepSRDCF [9] and CCOT [11]) or learning complex deep trackers (e.g. MDNet [28] and FCNT [33]). While such methods have been shown to work well, they suffer from high complexity: extracting deep features or applying deep tracking frameworks is very computationally expensive. This limits the real-time performance of such methods, even on high-end GPUs. This work proposes a Background-Aware CF based on hand-crafted features (HOG [6]) that can efficiently model how both the foreground and background of the object varies over time. Our approach, like conventional CFs, is extremely computationally efficient- and extensive experiments over multiple tracking benchmarks demonstrate the superior accuracy and real-time performance of our method compared to the state-of-the-art trackers.

Subfields: Computer Vision and Pattern Recognition, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/01c40508dcb6f8e9efcdefe49e22bc0ccaf8881c

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

In Global and Planetary Change

Citations in the corpus

TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild (2018, methodology): “Note that the maximum sequence length is 30, thus t ∈ [1 , 30].”
Visual Tracking via Adaptive Spatially-Regularized Correlation Filters (2019, methodology): “Usually, the CF-based trackers [18, 12, 11, 15, 8] exploit large numbers of cyclically shifted samples for learning, and convert the correlation operations in the spatial domain to the element-wise multiplications in the frequency domain, thereby reducing the computation complexity and improving…”
Online Multi-Expert Learning for Visual Tracking (2020, neutral): “[39] decouple the”
LaSOT: A High-Quality Benchmark for Large-Scale Single Object Tracking (2018, background): “…[9], SAMF [33], SCT4 [6], STC [57] and Staple [3]) or deep features ( e.g ., HCFT [37] and ECO [7]) and regularization techniques ( e.g ., BACF [15], SRDCF [10], CSRDCF [36], Staple CA [40] and STRCF [29]), ensemble trackers ( e.g ., PTAV [13], LCT [38], MEEM [56] and TLD [24]), sparse…”
Learning Target Candidate Association to Keep Track of What Not to Track (2021, background): “Discriminative appearance model based trackers [13, 3, 30, 33, 59, 16] aim to suppress distractors based on their appearance by integrating background information when learning the target classifier online.”
A Twofold Siamese Network for Real-Time Object Tracking (2018, methodology): “OTB benchmarks: We compare SA-Siam with BACF[17], PTAV[10], DSiamM[11], EAST[15], Staple[2] ,SiamFC[3], CFNet[38], LMCF[34] and LCT[27] on OTB 2013/50/100 benchmarks.”
Siamese Cascaded Region Proposal Networks for Real-Time Visual Tracking (2018, methodology): “We compare C-RPN to 35 approaches, including ECO [7], MDNet [34], SiamFC [2], VITAL [42], StructSiam [55], TRACA [6], BACF [13] and so forth.”
Transformer Meets Tracker: Exploiting Temporal Context for Robust Visual Tracking (2021, background): “Discriminative correlation filter (DCF) tackles the visual tracking by solving the ridge regression in Fourier domain, which exhibits attractive efficiency [20, 35, 34, 15, 37, 53, 11, 8].”
Distractor-aware Siamese Networks for Visual Object Tracking (2018, methodology): “Among the real-time trackers, SiamFC and CFNet are latest Siamese network based trackers while the accuracies is still left far behind the state-of-the-art BACF and ECO-HC with HOG features.”
Is First Person Vision Challenging for Object Tracking? (2020, neutral): “Specifically, we have considered short-term trackers [46] based on correlation-filters with hand-crafted features (MOSSE [11], DSST [23], KCF [37], Staple [7], BACF [42], DCFNet [79], STRCF [49], MCCTH [78]) and deep features (ECO [21], ATOM [22], DiMP [9], PrDiMP [24], KYS [10]).”
Multi-cue Correlation Filters for Robust Visual Tracking (2018, methodology): “enhanced by using multi-dimensional features [16, 26, 31], part-based strategies [29, 28], adaptive scale [9, 26], longterm framework [32, 17, 45], end-to-end learning [41, 40] and improved filter training methods [20, 30, 33, 14].”
AutoTrack: Towards High-Performance Visual Tracking for UAV With Automatic Spatio-Temporal Regularization (2020, methodology): “To improve DCF-based trackers, there are currently three directions: a) building more robust appearance model [18, 17, 20, 21], b) mitigating boundary effect or imposing restrictions in learning [8, 22, 17, 14, 23], and c) mitigating filter degradation [24, 12, 14, 25].”
The Seventh Visual Object Tracking VOT2019 Challenge Results (2019, methodology): “…a.varfolomieiev@kpi.ua The method is based on SRDCF formulation [17], which defers from the original one in two main points: 1) it calculates the ﬁlter channels for each feature channel independently, and 2) the regularization in the ﬁlter is performed iteratively using the ADMM approach [29].”
VITAL: VIsual Tracking via Adversarial Learning (2018, background): “Extensions include, but are not limited to, kernelized correlation filters [23], scale estimation [10], re-detection [37], spatial regularization [12, 14, 9], ADMM optimization [29], sparse representation [32, 42, 33], CNN feature integrations [36, 43, 64, 28] and end-to-end CNN predictions [57, 55,”
Learning Attentions: Residual Attentional Siamese Network for High Performance Online Visual Tracking (2018, background): “One of the most successful tracking frameworks is Correlation Filter (CF) [3,8,22,27,33,54,55].”
GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild (2018, methodology): “works on correlation filters based tracking: CSK [55], KCF [56] and their variants ColorCSK [62], SAMF [58], DSST [48], Staple [45], SRDCF [51], SRDCFdecon [52], CCOT [53], BACF [69], STRCF [81], DeepSTRCF [81], ECO [47], and LDES [82] in our evaluation.”