Learning Spatially Regularized Correlation Filters for Visual Tracking

By Martin Danelljan, Gustav Häger, Fahad Shahbaz Khan, Michael Felsberg. IEEE International Conference on Computer Vision, 2015

1,970 citations recorded by the source corpus.

Robust and accurate visual tracking is one of the most challenging computer vision problems. Due to the inherent lack of training data, a robust approach for constructing a target appearance model is crucial. Recently, discriminatively learned correlation filters (DCF) have been successfully applied to address this problem for tracking. These methods utilize a periodic assumption of the training samples to efficiently learn a classifier on all patches in the target neighborhood. However, the periodic assumption also introduces unwanted boundary effects, which severely degrade the quality of the tracking model. We propose Spatially Regularized Discriminative Correlation Filters (SRDCF) for tracking. A spatial regularization component is introduced in the learning to penalize correlation filter coefficients depending on their spatial location. Our SRDCF formulation allows the correlation filters to be learned on a significantly larger set of negative training samples, without corrupting the positive samples. We further propose an optimization strategy, based on the iterative Gauss-Seidel method, for efficient online learning of our SRDCF. Experiments are performed on four benchmark datasets: OTB-2013, ALOV++, OTB-2015, and VOT2014. Our approach achieves state-of-the-art results on all four datasets. On OTB-2013 and OTB-2015, we obtain an absolute gain of 8.0% and 8.2% respectively, in mean overlap precision, compared to the best existing trackers.

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

View Paper: https://www.semanticscholar.org/paper/09769e80cdf027db32a1fcb695a1aa0937214763

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

Unsupervised Deep Tracking (2019, neutral): “015, s = {−1, 0, 1}} following [10].”
The Visual Object Tracking VOT2017 Challenge Results (2017, methodology): “The top ten trackers according to the primary EAO measure (Figure 4) are LSART (A.24), CFWCR (A.14), CFCF (A.10), ECO (A.30), gnet (A.16), MCCT (A.4), CCOT (A.36), CSRDCF (A.38), SiamDCF (A.23), MCPF (A.18).”
RGB-T Object Tracking: Benchmark and Baseline (2018, methodology): “To address the boundary effect, SRDCF [52] is proposed by introducing spatial regularization into the learning of correlation filters to penalize the filter coefficients near the boundary.”
Learning Background-Aware Correlation Filters for Visual Tracking (2017, methodology): “Several recent works addressed the constraint of learning from shifted patches by exploiting training samples whose spatial size is much larger than the trained filters [17, 10, 8, 9].”
High Performance Visual Tracking with Siamese Region Proposal Network (2018, methodology): “In this experiment, we compare our method with several representive trackers, including PTAV [11], CREST[31], SRDCF [8], SINT [33], CSR-DCF [23], Siamese-FC [4], Staple [3], CFNet [35] and DSST [9].”
Learning Aberrance Repressed Correlation Filters for Real-Time UAV Tracking (2019, methodology): “Spatially regularized DCF (SRDCF) was proposed to introduce punishment for background in training correlation filters so that they can be learned in larger search regions [7].”
Deep Learning for Visual Tracking: A Comprehensive Survey (2019, background): “, DCF-based trackers) provide an appealing trade-off of competitive tracking performance and efficient computations [22]–[24].”
Adaptive Decontamination of the Training Set: A Unified Formulation for Discriminative Visual Tracking (2016, methodology): “Baseline [5] Baseline-Entropy Baseline-PSR Ours”
Learning Spatial-Temporal Regularized Correlation Filters for Visual Tracking (2018, methodology): “Along with the introduction of feature representation [14, 28], nonlinear kernel [19], scale estimation [11, 23, 24], max-margin classifiers [43], spatial regularization [13, 18], and continuous convolution [15], DCF-based trackers have been greatly improved and significantly advanced the state-of-t”
Learning to Track at 100 FPS with Deep Regression Networks (2016, background): “8 fps [26] to 15 fps [37], with the top performing trackers running at 1 fps on a GPU [30,7,21].”
Learning the Model Update for Siamese Trackers (2019, background): “Updating the object template Most trackers either use simple linear interpolation to update the template in every frame [3, 19, 9, 11, 20, 5] or do not update the initial template at all [1, 41, 27, 50].”
SiamCAR: Siamese Fully Convolutional Classification and Regression for Visual Tracking (2019, neutral): “43 CPU Matlab SRDCF [27] 0.”
Context-Aware Correlation Filter Tracking (2017, methodology): “In addition, we include popular CF trackers that did not meet the selection criteria for our framework, namely DSST [5] that is very similar to SAMF but usually underperforms it, MUSTER [12] that employs a long-term/shortterm memory strategy but its source code is not available for modification, and”
CREST: Convolutional Residual Learning for Visual Tracking (2017, methodology): “We also compare our CREST tracker on the OTB-2015 benchmark [46] with the 29 trackers in [45] and the state-of-the-art trackers, including KCF [17], MEEM [48], TGPR [13], DSST [7], MUSTer [20], LCT [31], HCFT [30], SRDCF [9], CNN-SVM [19], DeepSRDCF [8], Staple [2], SRDCFdecon [10], CCOT [11], HDT [”
Know Your Surroundings: Exploiting Scene Information for Object Tracking (2020, methodology): “A popular method to learn the target appearance model is the discriminative correlation ﬁlters (DCF) [5,21,10,9,26,33].”
A Benchmark and Simulator for UAV Tracking (2016, methodology): “In addition, we include several of the latest trackers such as MEEM [44], MUSTER [18], DSST [8] (winner VOT2014) and SRDCF [7] (winner VOT-TIR2015 and OpenCV challenge).”
Multi-task Correlation Particle Filter for Robust Object Tracking (2017, methodology): “art tracking methods including MEEM [39], TGPR [12], KCF [16], MUSTer [18], DSST [9], LCT [26], CF2 [25], HDT [29], Staple [2], SRDCF [10], DeepSRDCF [7], SRDCFdecon [8], CNN-SVM [17], and C-COT [11].”
Siamese Box Adaptive Network for Visual Tracking (2020, methodology): “We compare our tracker with other 9 state-of-art real-time trackers, including DiMP [2], ATOM [5], SiamRPN++ [20], DaSiamRPN [52], SiamRPN [21], ECO [6], ECO-HC [6], SRDCF [7], SAMF [24].”