Color attributes for object detection

By Fahad Shahbaz Khan, Rao Muhammad Anwer, Joost van de Weijer, Andrew D. Bagdanov, María Vanrell, Antonio M. López. 2012 IEEE Conference on Computer Vision and Pattern Recognition, 2012

321 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/0bca0ca7bb642b747797c17a4899206116fb0b25

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

Filtered channel features for pedestrian detection (2015, background): “Colour has shown to be an effective feature for pedestrian detection and hence multiple colour spaces have been explored (both hand-crafted and learned) [8, 17, 18, 22].”
NEIL: Extracting Visual Knowledge from Web Data (2013, methodology): “For the object and attribute section, we use CHOG [20] features with a bin size of 8.”
The Pascal Visual Object Classes Challenge: A Retrospective (2014, background): “UVA_MERGED · • · · University of Amsterdam Sezer Karaoglu, Fahad Khan, Koen van de Sande, Jan van Gemert, Rao Muhammad Anwer, Jasper Uijlings, Camp Davesa, Joost van de Weijer, Theo Gevers, Cees Snoek Khan et al. (2012a), Uijlings et al.”
Learning coarse-to-fine sparselets for efficient object detection and scene classification (2015, methodology): “state-of-the-art detection accuracy can be achieved if we use the proposed method to replace DPMs in [41, 42].”
Recognizing Actions Through Action-Specific Person Detection (2015, methodology): “1) Person Detection: Most state-of-the-art methods for person detection are based on the learning-fromexamples paradigm [6], [12], [25].”
Dynamic Match Kernel With Deep Convolutional Features for Image Retrieval (2018, methodology): “Local regions where both texture and color cues are sufficiently close are considered as a true match by incorporating two Kronecker delta function with SIFT and color name (CN) [7] descriptors.”
Enriching Visual Knowledge Bases via Object Discovery and Segmentation (2014, methodology): “Speciﬁcally, we train an exemplar detector [10, 24] (eLDA in our case) based on Color-HOG (CHOG) features [31].”
A Color-Texture-Structure Descriptor for High-Resolution Satellite Image Classification (2016, background): “Color-Texture-Structure Descriptor Generation”
Generic Object Detection with Dense Neural Patterns and Regionlets (2014, background): “Other architectures such as local receptive ﬁeld networks with untied weights (Le et al., 2012) or fully-connected networks 1 do not have these properties.”
A Scale Adaptive Kernel Correlation Filter Tracker with Feature Integration (2014, methodology): “As achieved the promising results in other visual tasks such as object recognition, object detection and action recognition [17] [19] [18], we employ the mapping method described in [32] to transform the RGB space into the color names space, which is an 11 dimensional color representation.”
SIFT Meets CNN: A Decade Survey of Instance Retrieval (2016, methodology): “ptical region of each SIFT descriptor in the CIE-Lab color space. It is binarized to combine with the SIFT binary codes [12]. A similar idea is reﬂected in [35] where the local Color Names descriptor [62] is extracted by coupling with SIFT. Both methods achieve competitive accuracy on several benchm”
ME R-CNN: Multi-Expert Region-based CNN for Object Detection (2017, methodology): “One of the conventional ways to generate RoIs is to use multi-scale sliding windows [2, 8, 11, 15, 21, 27, 35] which can be considered as a ‘dense’ search.”
Query-adaptive late fusion for image search and person re-identification (2015, background): “In early fusion, descriptors are combined at feature level [15, 27, 2] or even sensor level.”