Learning Color Names for Real-World Applications

By Joost van de Weijer, Cordelia Schmid, Jakob Verbeek, Diane Larlus. 2009

758 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

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

In Media Technology

Citations in the corpus

Network Dissection: Quantifying Interpretability of Deep Visual Representations (2017, methodology): “In addition, every image pixel in the data set is annotated with one of the eleven common color names according to the human perceptions classified by van de Weijer [31].”
Interpreting Deep Visual Representations via Network Dissection (2017, background): “of eleven color names commonly used by humans [37].”
Salient Color Names for Person Re-identification (2014, methodology): “Since color names show good robustness to photometric variance [26], an alternative approach is to apply color names to describe colors [12,27,10,26].”
Context-Aware Correlation Filter Tracking (2017, methodology): “Tracker Features Scale Published MOSSE grayscale pixel intensity No 2010 (CVPR) DCF HOG [8] No 2015 (TPAMI) SAMF HOG [8], CN [25] Yes 2014 (ECCV-W) STAPLE HOG [8], color histogram Yes 2016 (CVPR)”
Convolutional Features for Correlation Filter Based Visual Tracking (2015, methodology): “We compare with using grayscale intensity (I), Histogram of Oriented Grandients (HOG) [7] and Color Names (CN) [36].”
Nonparametric Part Transfer for Fine-Grained Recognition (2014, methodology): “Example images of the CUB-2011 dataset [36] with corresponding color name features as proposed by [32].”
Attribute-Based Classification for Zero-Shot Visual Object Categorization (background): “Prior work in the area of computer vision has mainly studied elementary properties, such as colors and geometric patterns [30], [31], [32], achieving high accuracy by developing taskspecific features and representations.”
Learning Spatially Regularized Correlation Filters for Visual Tracking (2015, background): “Recent work [9, 8, 10, 20, 24] have shown a notable improvement by learning multi-channel ﬁlters on multi-dimensional features, such as HOG [7] or Color-Names [31].”
ECO: Efficient Convolution Operators for Tracking (2016, methodology): “We apply the same feature representation as C-COT, namely a combination of the first (Conv-1) and last (Conv-5) convolutional layer in the VGG-m network [5], along with HOG [6] and Color Names (CN) [35].”
Local Novelty Detection in Multi-class Recognition Problems (2015, methodology): “As done in [36], we extract features from patches around ground truth part locations but restrict feature representations as in [13] to histograms over both quantized OpponentSIFT descriptors [31] and color name descriptors [38].”
Person Re-identification: Past, Present and Future (2016, methodology): “In [44], Zheng et al. propose extracting the 11-dim color names descriptor [45] for each local patch, and aggregating them into a global vector through a Bag-of-Words (BoW) model.”
Large-scale vehicle re-identification in urban surveillance videos (2016, methodology): “This is a benchmark method in person Re-Id, which applies BOW with Color Name (CN) feature [20].”
Online Multi-Object Tracking with Dual Matching Attention Networks (2018, methodology): “ine tracker. The ECO tracker achieves the state-of-the-art performance on visual tracking benchmarks [25,52,38,33] and its fast variant ECO-HC based on hand-crafted features (HOG [11] and Color Names [49]) operates at 60 frames per second (FPS) on a single CPU, which is suitable for the online MOT t”
Discriminative Scale Space Tracking (2016, positive): “intensitywith the Color Names representation [36].”
Deep learning in video multi-object tracking: A survey (2019, methodology): “They employed a Discriminative Correlation Filters (DCF) based online tracking method [56] with the use of HOG [57] and Colour Names [58] features to predict the position of so-called tag-boxes, small regions around the center of each animal.”
A Color-Texture-Structure Descriptor for High-Resolution Satellite Image Classification (2016, methodology): “High-dimensional color features, such as Hue [34] and CN [33], lead to high computational complexity, which reduces the efficiency of CBPT construction.”
Adaptive Color Attributes for Real-Time Visual Tracking (2014, methodology): “We use the mapping provided by [23], which is automatically learned from images retrieved with Google-image search.”