Visual object tracking using adaptive correlation filters

By David S. Bolme, J. Ross Beveridge, Bruce A. Draper, Yui Man Lui. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010

3,087 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/70c3c9b9a40ca55264e454586dca2a6cf416f6e0

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

MUlti-Store Tracker (MUSTer): A cognitive psychology inspired approach to object tracking (2015, background): “The correlation filter-based trackers [5, 6, 7, 17, 16, 31] are becoming increasingly popular due to their promising performance and computational efficiency.”
A Novel Performance Evaluation Methodology for Single-Target Trackers (2015, methodology): “The DSST, SAMF and KCF are correlation-filter-based trackers derived from MOSSE [75] that apply holistic models, i.e., a HOG [76].”
ECO: Efficient Convolution Operators for Tracking (2016, background): “The standard approach in DCF based tracking is to update the model in each frame [4, 9, 20].”
Understanding and Diagnosing Visual Tracking Systems (2015, methodology): “For example, fast Fourier transform (FFT) [5] and circular matrices [15, 9] are used to accelerate dense (kernelized) ridge regression.”
Convolutional Features for Correlation Filter Based Visual Tracking (2015, methodology): “As discussed in [4], the extracted features are always multiplied with a Hann window.”
Multi-task Correlation Particle Filter for Robust Object Tracking (2017, background): “Correlation filters have recently been introduced into visual tracking and shown to achieve high speed as well as robust performance [4, 9, 16, 15, 18, 26, 24, 21, 25, 29].”
Adaptive Masked Weight Imprinting for Few-Shot Segmentation (2019, methodology): “Similar to adaptive correlation filters (Bolme et al., 2010) that was used in visual object tracking with handcrafted features, the convolutional layer weights in our model can be updated with the newly imprinted weights for that class in an adaptive scheme.”
Need for Speed: A Benchmark for Higher Frame Rate Object Tracking (2017, methodology): “In recent years, significant attention has been paid to Correlation Filter (CF) based methods [3, 14, 9, 16, 17, 1] for visual tracking.”
Meta-Tracker: Fast and Robust Online Adaptation for Visual Object Trackers (2018, background): “From the early success of the MOSSE tracker [7], a large number of variations have been suggested.”
RGB-T Object Tracking: Benchmark and Baseline (2018, background): “[42], which achieves hundreds of frames per second and high tracking accuracy.”
FairMOT: On the Fairness of Detection and Re-identification in Multiple Object Tracking (2020, methodology): “We update the appearance features of the trackers in each time step to handle appearance variations as in [53], [54].”
CN-Celeb: A Challenging Chinese Speaker Recognition Dataset (2019, methodology): “We used the ArcFace face recognition system [18] to perform the comparison; (3) Apply the MOSSE face tracking system [19] to produce face streams.”
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].”
The Visual Object Tracking VOT2016 Challenge Results (2016, methodology): “The Discriminative Scale Space Tracker (DSST) [48] extends the Minimum Output Sum of Squared Errors (MOSSE) tracker [101] with robust scale estimation.”
Correlation filters with limited boundaries (2014, methodology): “The Peak-to-Sidelobe Ratio (PSR) [4] values show that our method returns stronger output compared to the other ﬁlters.”
Hedged Deep Tracking (2016, methodology): “[4] develop the Minimum Output Sum of Squared Error (MOSSE) method to learn the filters, and use intensity features for object representation.”
VITAL: VIsual Tracking via Adversarial Learning (2018, methodology): “Starting from the MOSSE tracker [5], many efforts have been made to improve the correlation filter for robust tracking.”
LaSOT: A High-quality Large-scale Single Object Tracking Benchmark (2020, methodology): “Owing to the extremely efficient solution using fast Fourier transform (FFT), correlation-filter trackers [8, 39] run at speeds of several hundred frames per second and draw extensive attention with many improvements.”
TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild (2018, methodology): “Also, the remarkable runtime performance is achieved by efficiently solving the underlying ridge regression problem in the Fourier domain [23].”