MEEM: Robust Tracking via Multiple Experts Using Entropy Minimization

By Jianming Zhang, Shugao Ma, Stan Sclaroff. 2014

1,129 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Learning feed-forward one-shot learners (2016, methodology): “An elementary tracker based on learnet compares favourably against recent tracking systems, which make use of different features and online model update strategies: DAT [17], DSST [2], MEEM [22], MUSTer [6] and SO-DLT [21].”
Learning Multi-domain Convolutional Neural Networks for Visual Tracking (2015, methodology): “…popular trends since it is difﬁcult to collect a large amount of training data for video processing applications and training algorithms specialized for visual tracking are not available yet, while the approaches based on low-level handcraft features still work well in practice [18, 6, 21, 42].”
Transferring Rich Feature Hierarchies for Robust Visual Tracking (2015, methodology): “Although there exist more advanced model update methods, e.g. [36], we ﬁnd that our simple scheme works quite well in practice.”
RGB-T Object Tracking: Benchmark and Baseline (2018, methodology): “To quantify the advantages of different tracking algorithms on RGB234, we present the attribute-based MPR and MSR scores in [49], C-COT [48], CF NET [39], CSR-DCF [53], SRDCF [52], SAMF [51], DSST [60], SOWP [11], KCF [44], MEEM [54], CSR [6], JSR [2] AND L1-PF [31].”
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 [”
Visual Object Tracking for Unmanned Aerial Vehicles: A Benchmark and New Motion Models (2017, methodology): “…(Possegger, Mauthner, and Bischof 2015) • a competitive particle filter based approach: HOGLR (Wang et al. 2015b) • an expert ensemble based approach: MEEM (Zhang, Ma, and Sclaroff 2014) • two deep learning based approaches: SO-DLT (Wang et al. 2015a), MDNet (Nam and Han 2016) All of these…”
Large Margin Object Tracking with Circulant Feature Maps (2017, methodology): “The ﬁrst one is to explore an effective tracking algorithm which can be designed to be either discriminative [17, 9, 13, 15, 22, 1, 31] or generative [24, 18, 16, 33] models.”
Beyond Correlation Filters: Learning Continuous Convolution Operators for Visual Tracking (2016, methodology): “DSST SAMF TGPR MEEM LCT HCF Staple SRDCF SRDCFdecon DeepSRDCF C-COT OTB-2015 60.6 64.7 54.0 63.4 70.1 65.5 69.9 72.9 76.7 77.3 82.4 Temple-Color 47.5 56.1 51.6 62.2 52.8 58.2 63.0 62.2 65.8 65.4 70.4 0 0.2 0.4 0.6 0.8 1 Overlap threshold 0 20 40 60 80 100 O v e rl a p P re c is io n [ % ] Success pl”
Modeling and Propagating CNNs in a Tree Structure for Visual Tracking (2016, methodology): “[38] presented a framework based on multiple snapshots of SVM-based trackers to recover from tracking failures.”
Triplet Loss in Siamese Network for Object Tracking (2018, methodology): “For reference, we also compare with recent trackers: DSST [5], MEEM [35], SAMF [19], and DLSSVM [21].”
TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild (2018, methodology): “We include CFNet [15] and SiameseFC [14] to represent CNN matching trackers and MEEM [21] and DLSSVM [50] for structured SVM based trackers.”
Adaptive Decontamination of the Training Set: A Unified Formulation for Discriminative Visual Tracking (2016, methodology): “Other methods use a combination of experts [16, 29], a separate tracking model [14, 22] or distance comparisons [9] for managing the training set.”
ATOM: Accurate Tracking by Overlap Maximization (2018, background): “This trend is clearly observed in the recent VOT2018 challenge [17], where older trackers such as KCF [13] and MEEM [37] still obtain competitive accuracy while exhibiting vastly inferior robustness.”
GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild (2018, methodology): “They include generative methods LK [82], IVT [71] and L1APG [72] and discriminative method MEEM [42].”
Learning Target Candidate Association to Keep Track of What Not to Track (2021, background): “Discriminative appearance model based trackers [12, 3, 27, 25, 51, 15] aim to suppress distractors based on their appearance by integrating background information when learning the target classifier online.”
LaSOT: A High-Quality Benchmark for Large-Scale Single Object Tracking (2018, methodology): “5 is that all the top seven trackers leverage deep feature, demon-Table B CN [11] CVPR14 DSST [8] BMVC14 MEEM [56] ECCV14 STC [57] ECCV14 SAMF [33] ECCVW14 LCT [38] CVPR15 SRDCF [10] ICCV15 HCFT [37] ICCV15 KCF [20] PAMI15 Staple [3] CVPR16 SINT [49] CVPR16 SCT4 [6] CVPR16 MDNet [42] CVPR16 SiamFC…”
Learning Dynamic Siamese Network for Visual Object Tracking (2017, background): “Our tracker, DSiamM, outperforms SiamFC, Staple, MEEM, DSST and GOTURN on all 11 attributes.”
ECO: Efficient Convolution Operators for Tracking (2016, methodology): “Instead of updating the model in a continuous fashion every frame, we use a sparser updating scheme, which is a common practice in non-DCF trackers [31, 38].”
Attentional Correlation Filter Network for Adaptive Visual Tracking (2017, methodology): “These datasets have been frequently used [3,8,15,16,18,29,42] as they include a large variety of environments to evaluate the general performance of visual trackers.”