Viewpoint Invariant Pedestrian Recognition with an Ensemble of Localized Features

By Douglas T. Gray, Hai Tao. European Conference on Computer Vision, 2008

1,617 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/d745cf8c51032996b5fee6b19e1b5321c14797eb

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

Person re-identification by probabilistic relative distance comparison (2011, methodology): “Related work–Although it has not been exploited for person re-identification, distance learning is a well-studied problem with a large number of methods reported in the literature [16, 5, 17, 7, 17, 12, 15, 9, 1].”
Part-Aligned Bilinear Representations for Person Re-identification (2018, background): “The early solutions of person re-identification mainly relied on hand-crafted features [36, 27, 18, 39], metric learning techniques [71, 73, 42, 28, 22, 26, 20], and probabilistic patch matching algorithms [6, 5, 48] to handle resolution/light/view/pose changes.”
End-to-End Deep Learning for Person Search (2016, background): “…works of person re-identification focus on manually designing discriminative features [35, 11, 37], automatically learning features with convolutional neural networks (CNN) [18, 1], learning feature transforms across camera views [25, 24, 30], and learning distance metrics [40, 10, 26, 23, 21].”
Deep Transfer Learning for Person Re-Identification (2016, background): “ an unsolved problem [67]. Earlier works focus on either designing view-insensitive feature representations [8,57,23,33,65,36], or learning an effective distance metric [56,34,28,3,61,63,35], or both [11,26]. Recently, inspired by the success of deep neural networks, particularly deep Convoluational”
Clothing Attributes Assisted Person Reidentification (2015, methodology): “The performance of the proposed method is approximately between those of ITM and RDC, and better than LMNN and Adaboost.”
Person Re-identification by Descriptive and Discriminative Classification (2011, methodology): “These approaches can mainly be subdivided in two groups: (a) methods which employ a representation of descriptive statistics of the human appearance (using hand crafted features) [4,6,17,22] and (b) approaches that are based on discriminative learning [3,8,13,16,18].”
Person re-identification by Local Maximal Occurrence representation and metric learning (2014, methodology): “40.00%, LOMO + XQDA 20.13%, ELF6 + XQDA 19.97%, HSV_Lab_LBP + XQDA 10.09%, gBiCov + XQDA (a) Cosine (b) XQDA Figure 5 .”
Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification (2017, background): “Hand-craft features [32, 14, 7, 33, 26, 51] can be directly employed for unsupervised”
Patch-based Knowledge Distillation for Lifelong Person Re-Identification (2022, background): “It consists of twelve ReID datasets, including five seen datasets (Market-1501 [73], CUHK-SYSU [64], DukeMTMC-reID [48], MSMT17_V2 [61] and CUHK03 [32]) for non-forgetting evaluation and seven unseen datasets (VIPeR [16], PRID [22], GRID [39], iLIDS [74], CUHK01 [31], CUHK02 [30] and SenseReID [70])”
Part-based clothing segmentation for person retrieval (2011, methodology): “Approaches can be divided into those based on biometric features such as facial features (e.g. [2, 3, 4, 8]), those based on clothing information alone (e.g. [5, 10, 11, 14, 18]), and hybrid approaches (e.g. [1, 9, 16, 17]).”
Reinforced Temporal Attention and Split-Rate Transfer for Depth-Based Person Re-identification (2017, methodology): “Existing methods of person re-identification typically focus on designing invariant and discriminant features [24,22,50,38,100,46,87,10,43], which can enable identification despite nuisance factors such as scale, location, partial occlusion and changing lighting conditions.”
Efficient PSD Constrained Asymmetric Metric Learning for Person Re-Identification (2015, background): “There are many other approaches beyond metric learning for person re-identification, such as the ensemble of local features (ELF) [6], SDALF [1], RankSVM [23], kBiCov [18], salience match [30], mid-level filter [32], local fisher discriminant analysis (LF) [22], to name a few.”
Lifelong Person Re-Identification via Adaptive Knowledge Accumulation (2021, methodology): “Specifically, we first merge VIPeR [6], PRID [8], GRID [25], i-LIDS [41], CUHK01 [17], CUHK02 [16], SenseReID [50] in accordance with their original train/test splits as a new benchmark.”
Deep feature learning with relative distance comparison for person re-identification (2015, methodology): “Zhao et al. [13] proposed the learning of mid-level features from hierarchical clusters of patches.”
Joint Learning of Single-Image and Cross-Image Representations for Person Re-identification (2016, methodology): “CUHK03 [19], CUHK01 [18] and VIPeR [12].”
Multi-Task Learning with Low Rank Attribute Embedding for Person Re-Identification (2015, methodology): “As for feature design, different kinds of features have been tailored and employed in previous work, including histogram features from various color and textture channels [15, 49], symmetry-driven accumulation of local features [12], features from body parts with pictorial structures [9] to estimate”