Fast Human Detection Using a Cascade of Histograms of Oriented Gradients

By Qiang Zhu, Mei-Chen Yeh, Kwang-Ting Cheng, Shai Avidan. Computer Vision and Pattern Recognition, 2006

575 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/05fe01b57b3ba58dc5029c068a48567b55018ea5

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 Electrical and Electronic Engineering

Citations in the corpus

Human Detection via Classification on Riemannian Manifolds (2007, methodology): “We would like to enhance this property to also include local illumination variations in an image.”
S^3FD: Single Shot Scale-Invariant Face Detector (2017, background): “After that, lots of works have focused on improving the performance with more sophisticated hand-crafted features [25, 28, 53, 60] and more powerful classiﬁers [2, 33].”
Pedestrian detection at 100 frames per second (2012, background): “By having the first stages prune most of the false positives, the average computation time is significantly reduced [18, 20, 10].”
AutoFocus: Efficient Multi-Scale Inference (2018, methodology): “Several methods since then have been proposed to improve it [6, 67, 65].”
Object Detection in 20 Years: A Survey (2019, unknown): “A typical example is to speed up HOG by computing integral HOG maps [124], [157], as shown in Fig.”
Fast Feature Pyramids for Object Detection (2014, methodology): “[63] proposed fast computation of gradient histograms using integral histograms [64]; the proposed system was real time for singlescale detection only.”
Recent advances and trends in visual tracking: A review (2011, background): “Zhu et al. [22] improved its computational efﬁciency signiﬁcantly by utilizing a boosted cascade of rejectors.”
Expecting the Unexpected: Training Detectors for Unusual Pedestrians with Adversarial Imposters (2017, background): “HOG+Cascade: A cascade of boosted classiﬁers working with HOG features [50].”
An HOG-LBP human detector with partial occlusion handling (2009, unknown): “While the integrated HOG feature without trilinear interpolation is fast to compute, it is inferior to the original HOG, as mentioned in [34].”
Cat Head Detection - How to Effectively Exploit Shape and Texture Features (2008, methodology): “[21] used HOGs of variable-size blocks in the boosting framework .”
Efficient Low-rank Multimodal Fusion With Modality-Specific Factors (2018, methodology): “Visual The library Facet1 is used to extract a set of visual features for each frame (sampled at 30Hz) including 20 facial action units, 68 facial landmarks, head pose, gaze tracking and HOG features (Zhu et al., 2006).”
Asynchronous parallel stochastic gradient descent: a numeric core for scalable distributed machine learning algorithms (2015, methodology): “In our case, large numbers of d = 128 dimensional HOG features [18] were extracted from a collection of images and clustered to form “Codebooks” with k = 100, .”
Multimodal Language Analysis with Recurrent Multistage Fusion (2018, unknown): “Visual: The library Facet (iMotions, 2017) is used to extract a set of visual features including facial action units, facial landmarks, head pose, gaze tracking and HOG features (Zhu et al., 2006).”
Social signal processing: Survey of an emerging domain (2009, methodology): “[34][223]) from local regions of the images (typically arranged in a regular grid), and then make a decision using classifiers like the Support Vector Machines.”
The Fastest Pedestrian Detector in the West (2010, methodology): “One of the earliest attempts to create real time detectors that utilize gradient histograms was the method of [36] (based on integral histograms [25]).”
Pedestrian Detection: An Evaluation of the State of the Art (2012, methodology): “Moreover, numerous approaches have been proposed for speeding up detection, including speeding up the detector itself [29], [44], [46], through use of approximations [63], [89] or by using special purpose hardware such as GPUs [90] (for a review of fast detection see [63]).”
Human Body Detection and Geolocalization for UAV Search and Rescue Missions Using Color and Thermal Imagery (2008, unknown): “The use of integral image representation and a rejection cascade allows for 5 to 30 Hz human detection performance (for images of 320x280 pixels size) [4].”
Objects in Action: An Approach for Combining Action Understanding and Object Perception (2007, methodology): “We employ a variant of the histogram of oriented gradient(HOG) approach from [5, 33].”