P-N learning: Bootstrapping binary classifiers by structural constraints

By Zdenek Kalal, Jiri Matas, Krystian Mikolajczyk. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010

1,180 citations recorded by the source corpus.

This paper shows that the performance of a binary classifier can be significantly improved by the processing of structured unlabeled data, i.e. data are structured if knowing the label of one example restricts the labeling of the others. We propose a novel paradigm for training a binary classifier from labeled and unlabeled examples that we call P-N learning. The learning process is guided by positive (P) and negative (N) constraints which restrict the labeling of the unlabeled set. P-N learning evaluates the classifier on the unlabeled data, identifies examples that have been classified in contradiction with structural constraints and augments the training set with the corrected samples in an iterative process. We propose a theory that formulates the conditions under which P-N learning guarantees improvement of the initial classifier and validate it on synthetic and real data. P-N learning is applied to the problem of on-line learning of object detector during tracking. We show that an accurate object detector can be learned from a single example and an unlabeled video sequence where the object may occur. The algorithm is compared with related approaches and state-of-the-art is achieved on a variety of objects (faces, pedestrians, cars, motorbikes and animals).

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/f2ea318fa9a6c56aac5aef133e16ce3307bb02af

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

Citations in the corpus

Forward-Backward Error: Automatic Detection of Tracking Failures (2010, methodology): “The FB error can also be implemented as a constraint and used in a novel tracking framework introduced in [6].”
Beyond Correlation Filters: Learning Continuous Convolution Operators for Visual Tracking (2016, methodology): “We validate our Continuous Convolution Operator Tracker (C-COT) in a comprehensive comparison with 20 state-of-the-art methods: ASLA [25], TLD [26], Struck [21], LSHT [22], EDFT [14], DFT [41], CFLB [18], ACT [13], TGPR [19], KCF [24], DSST [9], SAMF [31], MEEM [47], DAT [40], LCT [36], HCF [35], St”
Tracking-based semi-supervised learning (2011, methodology): “It is worth mentioning the related topic of discriminative tracking, e.g. Tang et al. (2007), Stalder et al. (2009) and Kalal et al. (2010), in which semi-supervised methods are sometimes used to address the model-free tracking problem.”
Memory Based Online Learning of Deep Representations from Video Streams (2017, background): “Only a few works on tracking have reported drift-free results on on very long video sequences ([50, 51, 52, 53, 54] among the few), and only few of them have provided convincing evidence on the possibility of incremental appearance learning strategies that are asymptotically stable [50][52].”
Adaptive Decontamination of the Training Set: A Unified Formulation for Discriminative Visual Tracking (2016, methodology): “EDFT[7] LSHT[12] DFT[24] ASLA[15] TLD[16] Struck[11] CFLB[8] ACT[6] TGPR[9] KCF[13] DSST[2] SAMF[19] DAT[23] MEEM[29] LCT[22] HCF[21] SRDCF[5] Ours OTB-2015 41.”
Tracking-Learning-Detection (2012, methodology): “In our previous work [59], we used only the first two stages.”
Learning Spatially Regularized Correlation Filters for Visual Tracking (2015, methodology): “We provide a comparison of our tracker with 24 state-of-the-art methods from the literature: MIL [2], IVT [28], CT [36], TLD [22], DFT [29], EDFT [12], ASLA [21], L1APG [3], CSK [19], SCM [37], LOT [26], CPF [27], CXT [11], Frag [1], Struck [16], LSHT [17], LSST [32], ACT [10], KCF [20], CFLB [14], ”
Adaptive Color Attributes for Real-Time Visual Tracking (2014, methodology): “CT [27] LSST [24] Frag [1] L1APG [2] LOT [18] ASLA [10] TLD [11] SCM [28] EDFT [6] CSK [9] DFT [20] CXT [5] CPF [19] LSHT [8] Struck [7] CN2 CN Median CLE 78.”
Convolutional Features for Correlation Filter Based Visual Tracking (2015, background): “[20], CFLB [16], MIL [3], CT [41], TLD [23], DFT [34], EDFT [15], ASLA [22].”
Learning a Deep Compact Image Representation for Visual Tracking (2013, background): “To address this problem, some trackers taking the semi-supervised learning approach have been proposed [12, 7].”
Robust Visual Tracking and Vehicle Classification via Sparse Representation (2011, neutral): “Examples include [2], [10], [46], [42], [3], and [23].”
Fast Visual Tracking via Dense Spatio-temporal Context Learning (2014, methodology): “Numerous algorithms have been proposed with focus on effective appearance models, which are based on the target appearance [8,1,28,22,17,18,19,23,21,31] or the difference between appearances of the target and its local background [11,16,14,2,30,15].”
ECO: Efficient Convolution Operators for Tracking (2016, methodology): “OTB2015 Dataset: We compare our tracker with 20 stateof-the-art methods: TLD [22], Struck [19], CFLB [16], ACT [13], TGPR [17], KCF [20], DSST [7], SAMF [25], MEEM [38], DAT [33], LCT [28], HCF [27], SRDCF [9], SRDCFad [10], DeepSRDCF [8], Staple [1], MDNet [31], SiameseFC [2], TCNN [30] and C-COT [”
Part-based multiple-person tracking with partial occlusion handling (2012, background): “On the other hand, discriminative tracking approaches with online learning have flourished recently [1, 2, 11, 6, 10, 8].”
Accurate Scale Estimation for Robust Visual Tracking (2014, methodology): “Ours [0.745] Struck [0.659] ASLA [0.612] SCM [0.610] TLD [0.509] LSHT [0.508] EDFT [0.505] CSK [0.502] L1APG [0.472] LOT [0.467] DFT [0.441] CT [ 0.344] 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 Overlap threshold Overlap Precision”