Staple: Complementary Learners for Real-Time Tracking

By Luca Bertinetto, Jack Valmadre, Stuart Golodetz, Ondrej Miksik, Philip H. S. Torr. Computer Vision and Pattern Recognition, 2015

1,648 citations recorded by the source corpus.

Correlation Filter-based trackers have recently achieved excellent performance, showing great robustness to challenging situations exhibiting motion blur and illumination changes. However, since the model that they learn depends strongly on the spatial layout of the tracked object, they are notoriously sensitive to deformation. Models based on colour statistics have complementary traits: they cope well with variation in shape, but suffer when illumination is not consistent throughout a sequence. Moreover, colour distributions alone can be insufficiently discriminative. In this paper, we show that a simple tracker combining complementary cues in a ridge regression framework can operate faster than 80 FPS and outperform not only all entries in the popular VOT14 competition, but also recent and far more sophisticated trackers according to multiple benchmarks.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/0f12a3aaf3851078d93a9bba4e3ebece6d4bcfe5

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

Kernel Cross-Correlator (2017, methodology): “Most importantly, KCC achieves better performance than Staple while running at roughly the same frame rate (40FPS), which clearly demonstrates the superiority.”
TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild (2018, methodology): “In recent years, correlation ﬁlter (CF) trackers [23–26] have emerged as the most common, fastest and most accurate category of trackers.”
Learning Dynamic Memory Networks for Object Tracking (2018, methodology): “We compare our method MemTrack with 9 recent real-time trackers (≥ 15 fps), including CFNet [30], LMCF [32], ACFN [5], RFL [36], SiamFC [4], SiamFC U [30], Staple [3], DSST [7], and KCF [15] on both OTB-2013 and OTB-2015.”
Learning Adaptive Discriminative Correlation Filters via Temporal Consistency Preserving Spatial Feature Selection for Robust Visual Object Tracking (2018, methodology): “Staple [42] employed colour histograms from foreground and background to generate a response map, which improved the reliability of the ﬁnal response.”
A Twofold Siamese Network for Real-Time Object Tracking (2018, methodology): “VOT2016 benchmark: We compare SA-Siam with the top real-time trackers in VOT2016 challenge [20] and Staple[2] , SiamFC[3] and ECOhc[6].”
Distractor-aware Siamese Networks for Visual Object Tracking (2018, methodology): “We evaluate the proposed algorithms with numerous fast and state-of-the-art trackers including SiamFC [2], CFNet [33], Staple [1], CSRDCF [19], BACF [11], ECO-HC [3], CREST [29], MDNet [23], CCOT [5], ECO [3], and the baseline tracker SiamRPN [16].”
Fully-Convolutional Siamese Networks for Object Tracking (2016, methodology): “We also include Staple [33] and GOTURN [28], two recent realtime trackers presented respectively at CVPR 2016 and ECCV 2016.”
The Visual Object Tracking VOT2016 Challenge Results (2016, methodology): “For more details, we refer the reader to [60].”
Online Multi-Expert Learning for Visual Tracking (2020, neutral): “Not only using Histogram of Oriented Gradient(HOG) as the feature, researchers proposed a method named Staple [32] to take color information into account and achieved high performance.”
Triplet Loss in Siamese Network for Object Tracking (2018, methodology): “On OTB-2013 [32], OTB-50, and OTB-100 [33] benchmarks, we compare improved trackers: SiamFC-tri, CFnet2-tri and SiamImp-tri against several state-of-the-art real-time trackers: SiamFC [2], CFnet2, SiamImp [28], Staple [1], CN [5], and KCF [12].”
AutoTrack: Towards High-Performance Visual Tracking for UAV With Automatic Spatio-Temporal Regularization (2020, background): “There are currently two main research interests in this area: discriminative correlation filter (DCF)-based methods [5, 6, 7, 8, 9, 10, 11, 12, 13, 14] as well as deep learningbased approaches [15, 16, 17, 18, 19, 20].”
VITAL: VIsual Tracking via Adversarial Learning (2018, methodology): “We compare VITAL with 29 trackers from the OTB-2013 benchmark [59] and other 28 stateof-the-art trackers including DSST [10], KCF [22], TGPR [16], MEEM [63], RPT [34], LCT [37], MUSTer [25], HCFT [36], FCNT [57], SRDCF [12], CNN-SVM [24], DeepSRDCF [11], DAT [41], Staple [3], SRDCFdecon [13], CCOT [”
Is First Person Vision Challenging for Object Tracking? (2020, positive): “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]).”
Unveiling the Power of Deep Tracking (2018, methodology): “We compare our method with the following state-of-the-art methods: CCOT [7], ECO [1], MDNet [13], HDT [20], FCNT [12], SRDCF [18], DeepSRDCF [40], BACF [41], Staple [3], MEEM [4], HCF [8], SAMF [42], DSST [43], MUSTER [25], Struck [44], and ASLA [45].”
ECO: Efficient Convolution Operators for Tracking (2016, methodology): “Our ECO-HC tracker also operates in real-time (60 FPS), with an AUC score of 51.7%, significantly outperforming Staple by 6.4%.”
End-to-End Representation Learning for Correlation Filter Based Tracking (2017, methodology): “We compare our methods against state-of-the-art trackers that can operate in realtime: SiamFC-3s [3], Staple [2] and LCT [23].”
Learning Background-Aware Correlation Filters for Visual Tracking (2017, methodology): “We extensively evaluate our tracker on four standard datasets, including OTB50 [36], OTB100 [37], TempleColor128 (TC128) [23], and VOT2015 [18], comparing with 24 state-of-the-art methods, such as TLD [15], Struck [12], CFLB [17], KCF [13], DSST [7], SAMF [22], MEEM [38], DAT [29], LCT [27], HCF [26”