Incremental Learning for Robust Visual Tracking

By David A. Ross, Jongwoo Lim, Ruei-Sung Lin, Ming-Hsuan Yang. International Journal of Computer Vision, 2008

3,290 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/505f48d8236eb25f871da272c2ac2fe4b41ea289

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

Learning Multi-domain Convolutional Neural Networks for Visual Tracking (2015, methodology): “Various generative target appearance modeling algorithms have been proposed including sparse representation [31, 43], density estimation [16, 22], and incremental subspace learning [32].”
Object Tracking Benchmark (2015, methodology): “Effective update algorithms have also been proposed via online mixture model [29], online boosting [22], and incremental subspace update [47].”
Consistent video depth estimation (2020, methodology): “Learning on testing data has been used in several different problem contexts: online update in visual tracking [Kalal et al. 2011; Ross et al. 2008], adapting object detectors from images to videos [Jain and Learned-Miller 2011; Tang et al.”
Face tracking and recognition with visual constraints in real-world videos (2008, methodology): “One reason for this is that IVT lacks strong mechanisms for detection or correction of drifting.”
Robust Object Tracking via Sparse Collaborative Appearance Model (2014, methodology): “The IVT method [4] tracks the target well as the subspace learning method is robust to illumination changes and small pose”
Video Processing From Electro-Optical Sensors for Object Detection and Tracking in a Maritime Environment: A Survey (2016, methodology): “Another method is the use of principal component analysis, where a low-dimensional subset of principal components is updated as new frames arrive [199], [200].”
Graph mode-based contextual kernels for robust SVM tracking (2011, methodology): “Using the polluted training data, both VTD and IPCA may learn an inaccurate or wrong subspace model for object representation, leading to inaccurate or wrong tracking results.”
Visual object tracking using adaptive correlation filters (2010, methodology): “Recent examples include: Incremental Visual Tracking (IVT) [17], Robust Fragmentsbased Tracking (FragTrack) [1], Graph Based Discriminative Learning (GBDL) [19], and Multiple Instance Learning (MILTrack) [2].”
Robust visual tracking using ℓ1 minimization (2009, positive): “In [28], the authors present a tracking method that incrementally learns a low-dimensional subspace representation, efficiently adapting online to changes in the appearance of the target.”
Robust Object Tracking with Online Multiple Instance Learning (2011, positive): “…s of the tracker location at time ðt À 1Þ: pð' Ã t j' Ã tÀ1 Þ / 1 if ' Ã t À ' Ã tÀ1 < s 0 otherwise: & ð2Þ This could be extended with something more sophisticated, such as a particle filter, as is done in [12], [29], [36]; however, we again emphasize that our focus is on the appearance model.”
Hierarchical Convolutional Features for Visual Tracking (2015, methodology): “Numerous hand-crafted features have been used to represent the target appearance such as subspace representation [24] and color histograms [37].”
Recent advances and trends in visual tracking: A review (2011, methodology): “In Ross et al. [73], proposed a generalized visual tracking framework based on the incremental image-as-vector subspace learning method with a sample mean update.”
Tracking-Learning-Detection (2012, background): “6.1 Comparison 1: CoGD TLD was compared with results reported in [33] which reports on performance of five trackers (Iterative Visual Tracking (IVT) [22], Online Discriminative Features (ODV) [27], Ensemble Tracking (ET) [28], Multiple Instance Learning (MIL) [30], and Cotrained Generative-Discrimin”
Visual Tracking: an Experimental Survey (2013, neutral): “IVT [55] Bounding box Array, intensities Gaussian Subspace match array Incremental extended model by iPCA”
Robust Visual Tracking using 1 Minimization (2009, positive): “In [28], the authors present a tracking method that incrementally learns a low-dimensional subspace representation, efficiently adapting online to changes in the appearance of the target.”