PROST: Parallel robust online simple tracking

By Jakob Santner, Christian Leistner, Amir Saffari, Thomas Pock, Horst Bischof. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010

430 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/16e36a4b59e214786737aa4ebc3ba86075b61e49

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

A Review on Deep Learning Techniques for Video Prediction (2020, methodology): “PROST [132] 2010 R 4 (10) 4936 (9296) All (bb) variable - X 7 7 7 7 Object bb I Arcade Learning [133] 2013 S custom3 custom3 0 210× 160 - X 7 7 7 7 7 -”
Long-term correlation tracking (2015, methodology): “The first issue is the well-known stability-plasticity dilemma [16, 20].”
Graph mode-based contextual kernels for robust SVM tracking (2011, background): “In contrast, discriminative learning based appearance models (DLMs) aim to maximize the inter-class separability between the object and non-object regions using discriminative learning techniques, including SVMs [7, 8, 21], boosting [5, 6], random forest [14], multiple instance learning [9], spatial”
Self-Paced Learning for Long-Term Tracking (2013, background): “Mean Center Displacement Error (MCDE) Tracker Us[C6] Us[C7] ART[24] LRSVT[25] TLD MIL[7] Frag[30] PROST[31] NN[32] OAB1[33]”
Object Tracking Benchmark (2015, background): “Sant-ner et al. [48] proposed an approach that combines static, moderately adaptive and highly adaptive trackers to account for appearance changes.”
Online Object Tracking: A Benchmark (2013, background): “[48] proposed an approach that combines static, moderately adaptive and highly adaptive trackers to account for appearance changes.”
A survey of appearance models in visual object tracking (2013, neutral): “In particular, we additionally address several interesting issues for the future research in Section 6.”
Hough Forests for Object Detection, Tracking, and Action Recognition (2011, background): “The resulting single detector is not only more convenient, but also achieves higher accuracy in multi-aspect detection tasks.”
Robust Superpixel Tracking (2014, methodology): “The proposed superpixel-based tracking (SPT) algorithm is evaluated against several state-of-the-art tracking methods, including the IVT [3], Frag [5], MIL [7], 1 [11], PROST [18], VTD [12], TLD [13], Struck [14] and HT [32] tracking methods.”
Robust tracking using local sparse appearance model and K-selection (2011, methodology): “PROST Sequences: To further evaluate our SPT method for accurate tracking under occlusion, appearance blur, and pose variation, the latest four sequences provided in [20] are selected to compare with PROST, MIL, and FragTracker [1].”
Robust Object Tracking via Sparse Collaborative Appearance Model (2014, background): “This update scheme preserves the first template ψ f (which is usually correct [10], [23], [34]) and takes the most recent tracking result into account.”
Real-Time Compressive Tracking (2012, background): “The Animal, Shaking and Soccer sequences are provided in [28] and the Box and Jumping are from [29].”
Understanding and Diagnosing Visual Tracking Systems (2015, methodology): “first noticed this issue, and then proposed a simple yet effective method [27] to balance the stability and plasticity by combining the results of template matching, random forest and optical flow.”
Visual tracking via adaptive structural local sparse appearance model (2012, methodology): “We evaluate the performance of the proposed algorithm on nine challenging sequences from prior work [1, 3, 10, 18, 19], the CAVIAR data set (http://groups.”
Distribution fields for tracking (2012, methodology): “In PROST, three different trackers are used and combined using a cascade.”
Superpixel tracking (2011, methodology): “The PROST method [20] extends the tracking-by-detection framework with multiple modules for reducing drifts.”
Tracking-Learning-Detection (2012, methodology): “TLD scored best in 9/10, outperforming by more than 12 percent the second best (Prost [67]).”