A Multilevel Mixture-of-Experts Framework for Pedestrian Classification
IEEE Transactions on Image ProcessingPublished 12 April 2011
Markus Enzweiler, Dariu M. Gavrila
Citations142
SJR quartileQ1
SJR score2.50
SNIP3.41
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
This work presents a novel multilevel Mixture-of-Experts approach to combine information from multiple features and cues with the objective of improved pedestrian classification, and shows a significant performance boost.
Abstract
Notwithstanding many years of progress, pedestrian recognition is still a difficult but important problem. We present a novel multilevel Mixture-of-Experts approach to combine information from multiple features and cues with the objective of improved pedestrian classification. On pose-level, shape cues based on Chamfer shape matching provide sample-dependent priors for a certain pedestrian view. On modality-level, we represent each data sample in terms of image intensity, (dense) depth, and (dense) flow. On feature-level, we consider histograms of oriented gradients (HOG) and local binary patterns (LBP). Multilayer perceptrons (MLP) and linear support vector machines (linSVM) are used as expert classifiers.
Keywords
Computer Science
International Journal of Computer VisionDistinctive Image Features from Scale-Invariant Keypoints
55,266 Citations2004David Lowe
This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene and can robustly identify objects among clutter and occlusion while achieving near real-time performance.
Histograms of Oriented Gradients for Human Detection
31,896 Citations2005Navneet Dalal, Bill Triggs
It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.
IEEE Transactions on Pattern Analysis and Machine IntelligenceObject Detection with Discriminatively Trained Part-Based Models
9,976 Citations2009Pedro F. Felzenszwalb, Ross Girshick +2 more
An object detection system based on mixtures of multiscale deformable part models that is able to represent highly variable object classes and achieves state-of-the-art results in the PASCAL object detection challenges is described.
Advances in industrial controlPattern Classification
7,884 Citations2000Evan L. Russell, Leo H. Chiang +1 more
Pattern RecognitionA comparative study of texture measures with classification based on featured distributions
6,753 Citations1996Timo Ojala, Matti Pietikäinen +1 more
This paper evaluates the performance both of some texture measures which have been successfully used in various applications and of some new promising approaches proposed recently.
IEEE Transactions on Pattern Analysis and Machine IntelligenceStatistical pattern recognition: a review
6,688 Citations2000Anil K. Jain, Peter Duin +1 more
The objective of this review paper is to summarize and compare some of the well-known methods used in various stages of a pattern recognition system and identify research topics and applications which are at the forefront of this exciting and challenging field.
LIBLINEAR: A Library for Large Linear Classification
6,625 Citations2008Rong-En Fan, Kai‐Wei Chang +3 more
IEEE Transactions on Pattern Analysis and Machine IntelligenceOn combining classifiers
5,227 Citations1998Josef Kittler, M. Hatef +2 more
Neural ComputationAdaptive Mixtures of Local Experts
4,988 Citations1991Robert A. Jacobs, Michael I. Jordan +2 more
A new supervised learning procedure for systems composed of many separate networks, each of which learns to handle a subset of the complete set of training cases, which is demonstrated to be able to be solved by a very simple expert network.
Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods
4,863 Citations1999John Platt
The output of a lassi(cid:12)er should be a alibrated posterior probability to enable post-pro essing and a method to train a kernel lassi with a logit link and a regularized maximum likelihood is proposed.
IEEE Transactions on Pattern Analysis and Machine IntelligenceStereo Processing by Semiglobal Matching and Mutual Information
4,147 Citations2007Heiko Hirschmüller
Pattern Classification (2nd ed.)
2,513 Citations1999Richard O. Duda, Peter E. Hart +1 more
Computer Vision and Image UnderstandingA survey of advances in vision-based human motion capture and analysis
2,508 Citations2006Thomas B. Moeslund, Adrian Hilton +1 more
This survey reviews recent trends in video-based human capture and analysis, as well as discussing open problems for future research to achieve automatic visual analysis of human movement.
International Journal of Computer VisionDetecting Pedestrians Using Patterns of Motion and Appearance
2,104 Citations2005Paul Viola, Michael Jones +1 more
This pedestrian detection system is the first to combine both sources of information in a single detector, and operates on low resolution images under difficult conditions (such as rain and snow).
Computer Vision and Image UnderstandingThe Visual Analysis of Human Movement: A Survey
1,767 Citations1999Dariu M. Gavrila
A number of promising applications are identified and an overview of recent developments in this domain is provided, including work on whole-body or hand motion and the various methodologies.
An HOG-LBP human detector with partial occlusion handling
1,688 Citations2009Xiaoyu Wang, Tony Xiao Han +1 more
By combining Histograms of Oriented Gradients (HOG) and Local Binary Pattern (LBP) as the feature set, this work proposes a novel human detection approach capable of handling partial occlusion and achieves the best human detection performance on the INRIA dataset.
Lecture notes in computer scienceHuman Detection Using Oriented Histograms of Flow and Appearance
1,653 Citations2006Navneet Dalal, Bill Triggs +1 more
A detector for standing and moving people in videos with possibly moving cameras and backgrounds is developed, testing several different motion coding schemes and showing empirically that orientated histograms of differential optical flow give the best overall performance.
Fast Human Detection Using a Cascade of Histograms of Oriented Gradients
1,398 Citations2006Qiang Zhu, Mei-Chen Yeh +2 more
This work integrates the cascade-of-rejectors approach with the Histograms of Oriented Gradients (HoG) features to achieve a fast and accurate human detection system that can process up to 30 frames per second while maintaining an accuracy level similar to existing methods.
2009 IEEE Conference on Computer Vision and Pattern RecognitionPedestrian detection: A benchmark
1,327 Citations2009Piotr Dollár, Christian Wojek +2 more
The Caltech Pedestrian Dataset is introduced, which is two orders of magnitude larger than existing datasets and proposes improved evaluation metrics, demonstrating that commonly used per-window measures are flawed and can fail to predict performance on full images.
International Journal of Computer VisionA Trainable System for Object Detection
1,282 Citations2000Constantine Papageorgiou, Tomaso Poggio
IEEE Transactions on Pattern Analysis and Machine IntelligenceMonocular Pedestrian Detection: Survey and Experiments
1,217 Citations2008Markus Enzweiler, Dariu M. Gavrila
An overview of the current state of the art of pedestrian detection from both methodological and experimental perspectives is provided and a clear advantage of HOG/linSVM at higher image resolutions and lower processing speeds is indicated.
IEEE Transactions on Pattern Analysis and Machine IntelligenceSurvey of Pedestrian Detection for Advanced Driver Assistance Systems
962 Citations2009David Gerónimo, Antonio M. López +2 more
This work divides the problem of detecting pedestrians from images into different processing steps, each with attached responsibilities, and separates the different proposed methods with respect to each processing stage, favoring a comparative viewpoint.
IEEE Transactions on Pattern Analysis and Machine IntelligenceExample-based object detection in images by components
945 Citations2001Anand Mohan, C. Papageorgiou +1 more
Results suggest that the improvement in performance is due to the component-based approach and the ACC data classification architecture, which is capable of locating partially occluded views of people and people whose body parts have little contrast with the background.
Classification using intersection kernel support vector machines is efficient
937 Citations2008Subhransu Maji, Alexander C. Berg +1 more
It is shown that one can build histogram intersection kernel SVMs (IKSVMs) with runtime complexity of the classifier logarithmic in the number of support vectors as opposed to linear for the standard approach.
Pedestrian Detection in Crowded Scenes
864 Citations2005Bastian Leibe, Edgar Seemann +1 more
Qualitative and quantitative results on a large data set confirm that the core part of the method is the combination of local and global cues via probabilistic top-down segmentation that allows examining and comparing object hypotheses with high precision down to the pixel level.
IEEE Transactions on Pattern Analysis and Machine IntelligenceLearning to detect objects in images via a sparse, part-based representation
847 Citations2004Sugandha Agarwal, Abdullah Bin Younas Awan +1 more
A learning-based approach to the problem of detecting objects in still, gray-scale images that makes use of a sparse, part-based representation is developed and a critical evaluation of the approach under the proposed standards is presented.
Multiple kernels for object detection
766 Citations2009Andrea Vedaldi, Varun Gulshan +2 more
This work uses multiple kernel learning of Varma and Ray (ICCV 2007) to learn an optimal combination of exponential χ2 kernels, each of which captures a different feature channel.
International Journal of Computer VisionDetection and Tracking of Multiple, Partially Occluded Humans by Bayesian Combination of Edgelet based Part Detectors
763 Citations2007Bo Wu, Ram Nevatia
This work presents an approach to automatically detect and track multiple, possibly partially occluded humans in a walking or standing pose from a single camera, which may be stationary or moving.
Lecture notes in computer scienceHuman Detection Based on a Probabilistic Assembly of Robust Part Detectors
610 Citations2004Krystian Mikolajczyk, Cordelia Schmid +1 more
A novel method for human detection in single images which can detect full bodies as well as close-up views in the presence of clutter and occlusion is described.
IEEE Transactions on Pattern Analysis and Machine IntelligenceAn Experimental Study on Pedestrian Classification
554 Citations2006Stefan Munder, Dariu M. Gavrila
An in-depth experimental study on pedestrian classification; multiple feature-classifier combinations are examined with respect to their ROC performance and efficiency and show that the novel combination of SVMs with LRF features performs best.
IEEE Transactions on Intelligent Transportation SystemsPedestrian Protection Systems: Issues, Survey, and Challenges
541 Citations2007T. Gandhi, Mohan M. Trivedi
A comprehensive review of research efforts underway dealing with pedestrian safety and collision avoidance is presented, including research dealing with probabilistic modeling of pedestrian behavior for predicting collisions between pedestrians and vehicles.
Depth and Appearance for Mobile Scene Analysis
535 Citations2007Andreas Ess, Bastian Leibe +1 more
International Journal of Computer VisionMulti-cue Pedestrian Detection and Tracking from a Moving Vehicle
522 Citations2006Dariu M. Gavrila, Stefan Munder
This paper presents a multi-cue vision system for the real-time detection and tracking of pedestrians from a moving vehicle, with results from extensive field tests in difficult urban traffic conditions suggest system performance is at the leading edge.
Human detection using partial least squares analysis
490 Citations2009William Robson Schwartz, Aniruddha Kembhavi +2 more
This paper describes a human detection method that augments widely used edge-based features with texture and color information, providing us with a much richer descriptor set, and is shown to outperform state-of-the-art techniques on three varied datasets.
Human Detection via Classification on Riemannian Manifolds
464 Citations2007Oncel Tuzel, Fatih Porikli +1 more
A novel approach for classifying points lying on a Riemannian manifold by incorporating the a priori information about the geometry of the space.
New features and insights for pedestrian detection
458 Citations2010Stefan Walk, Nikodem Majer +2 more
This work shows that motion features derived from optic flow yield substantial improvements on image sequences, if implemented correctly — even in the case of low-quality video and consequently degraded flow fields, and introduces a new feature, self-similarity on color channels, which consistently improves detection performance across different datasets.
2009 IEEE Conference on Computer Vision and Pattern RecognitionMulti-cue onboard pedestrian detection
351 Citations2009Christian Wojek, Stefan Walk +1 more
Evaluating different features and classifiers in a sliding-window framework indicates that incorporating motion information improves detection performance significantly and the combination of multiple and complementary feature types can also help improve performance.
Pedestrian detection for driving assistance systems: single-frame classification and system level performance
348 Citations2004Amnon Shashua, Yoram Gdalyahu +1 more
The functional and architectural breakdown of a monocular pedestrian detection system is described and the approach for single-frame classification based on a novel scheme of breaking down the class variability by repeatedly training a set of relatively simple classifiers on clusters of the training set is described.
Dynamic 3D Scene Analysis from a Moving Vehicle
297 Citations2007Bastian Leibe, Nico Cornelis +2 more
A system that integrates fully automatic scene geometry estimation, 2D object detection, 3D localization, trajectory estimation, and tracking for dynamic scene interpretation from a moving vehicle and demonstrates the performance of this integrated system on challenging real-world data showing car passages through crowded city areas.
IEEE Transactions on Pattern Analysis and Machine IntelligenceA Bayesian, Exemplar-Based Approach to Hierarchical Shape Matching
265 Citations2007Dariu M. Gavrila
A Bayesian model to estimate the a posteriori probability of the object class, after a certain match at a node of the tree, is presented, takes into account object scale and saliency and allows for a principled setting of the matching thresholds such that unpromising paths in the tree traversal process are eliminated early on.
Multi-cue pedestrian classification with partial occlusion handling
233 Citations2010Markus Enzweiler, Angela Eigenstetter +2 more
In experiments on extensive real-world data sets, with both partially occluded and non-occluded pedestrians, significant performance boosts over state-of-the-art approaches are obtained by up to a factor of four in reduction of false positives at constant detection rates.
Structure- and motion-adaptive regularization for high accuracy optic flow
174 Citations2009Andreas Wedel, Daniel Cremers +2 more
Lecture notes in computer scienceMultiple Component Learning for Object Detection
140 Citations2008Piotr Dollár, Boris Babenko +3 more
The method, Multiple Component Learning (mcl), automatically learns individual component classifiers and combines these into an overall classifier, and unlike methods that are not part-based, mcl is quite robust to occlusions.
Real-time detection of moving objects from moving vehicles using dense stereo and optical flow
134 Citations2005A. Talukder, Larry Matthies
This work describes algorithms to robustly estimate 6-DOF robot egomotion in the presence of moving objects using dense flow and dense stereo, and uses these estimates to identify other moving objects while the robot itself is moving.
Optimizing discrimination-efficiency tradeoff in integrating heterogeneous local features for object detection
128 Citations2008Bo Wu, Ram Nevatia
A framework to optimize the discrimination-efficiency tradeoff in integrating multiple, heterogeneous features for object detection and shows that this approach outperforms the state-of-the-art methods.
IEEE Transactions on Intelligent Transportation SystemsStereo-Based Pedestrian Detection for Collision-Avoidance Applications
114 Citations2009Sergiu Nedevschi, Silviu Bota +1 more
This work presents a new approach for standing- and walking-pedestrian detection, in urban traffic conditions, using grayscale stereo cameras mounted on board a vehicle, using pattern matching and motion for pedestrian detection.
Integrated pedestrian classification and orientation estimation
81 Citations2010Markus Enzweiler, Dariu M. Gavrila
This paper presents a novel approach to single-frame pedestrian classification and orientation estimation in terms of a set of view-related models which couple discriminative expert classifiers with sample-dependent priors, facilitating easy integration of other cues in a Bayesian fashion.
The International Journal of Robotics ResearchA Fast Stereo-based System for Detecting and Tracking Pedestrians from a Moving Vehicle
76 Citations2009Max Bajracharya, Baback Moghaddam +3 more
A fully integrated system for detecting, localizing, and tracking pedestrians from a moving vehicle that can reliably detect upright pedestrians to a range of 40 m in lightly cluttered urban environments and on a diverse set of datasets with groundtruth in outdoor environments with varying degrees of pedestrian density and clutter.
Lecture notes in computer scienceSupport Vector Machines with Example Dependent Costs
70 Citations2003Ulf Brefeld, Peter Geibel +1 more
A natural cost-sensitive extension of the support vector machine (SVM) is presented and its relation to the Bayes rule is discussed and an approach for including example dependent costs into an arbitrary cost-insensitive learning algorithm is derived by sampling according to modified probability distributions.
Journal of NeuroscienceHuman Cortical Object Recognition from a Visual Motion Flowfield
69 Citations2003Nikolaus Kriegeskorte, Bettina Sorger +5 more
The human motion complex appears to have a central role in SFM object recognition, not merely representing the motion flowfield but also the surface structure of the motion-defined object.
Lecture notes in computer scienceComputer Vision – ECCV 2010
64 Citations2010Stefan Walk, Konrad Schindler +1 more
A simple yet highly effective novel feature based on binocular disparity is proposed, outperforming previously proposed stereo features and it is shown that the combination of different classifiers often improves performance even when classifiers are based on the same feature or feature combination.
IEEE Transactions on Intelligent Transportation SystemsOn Exploration of Classifier Ensemble Synergism in Pedestrian Detection
58 Citations2009Luciano Oliveira, Urbano Nunes +1 more
This paper addresses the problem of pedestrian detection by a novel ensemble method, and analyzed the following two classes of fusion methods of combining the outputs of the component classifiers: majority vote and fuzzy integral.
Detection and Tracking of Multiple Humans with Extensive Pose Articulation
55 Citations2007Li Zhang, Bo Wu +1 more
This work describes a method for detecting and tracking humans with extensive pose articulations, under situations where there is typically only a single camera, multiple humans are present and the image resolution is low.
Towards Robust Pedestrian Detection in Crowded Image Sequences
52 Citations2007Edgar Seemann, Mario Fritz +1 more
This work presents a generative object model that is capable to scale from a general object class model to a more specific object-instance model that allows to detect class instances as well as to distinguish between individual object instances reliably.
IEEE Transactions on Intelligent Transportation SystemsA Comprehensive Evaluation Framework and a Comparative Study for Human Detectors
49 Citations2009Mohamed E. Hussein, Fatih Porikli +1 more
A framework for evaluating human detectors that considers the practical application of a detector on a full image using multisize sliding-window scanning suggests that the descriptors used to represent features and the training window size are more important in predicting the detection performance than the nature of the imaging process.
Monocular pedestrian recognition using motion parallax
43 Citations2008Markus Enzweiler, Patsy Kanter +1 more
It is demonstrated that the proposed focus-of-attention strategy reduces the false positives of an otherwise identical monocular pedestrian recognition system by a factor of two, at equal detection rates.
Image and Vision ComputingA time delay neural network algorithm for estimating image-pattern shape and motion
36 Citations1999Christian Wöhler, Joachim K. Anlauf
A novel concept for simultaneous shape estimation and motion analysis based on a feed-forward TDNN architecture with adaptable spatio-temporal receptive fields that turns out to be rather robust with respect to random deviations of the actual motion from the trained motion patterns.
A multiple detector approach to low-resolution FIR pedestrian recognition
32 Citations2005Mirko Mählisch, Matthias Oberländer +3 more
A recognition scheme, which is both reliable and fast, which comprises the simultaneous harmonized use of three powerful detection algorithms, the hyper permutation network (HPN), a hierarchical contour matching (HCM) algorithm and a cascaded classifier approach.
Lecture notes in computer scienceDisparity Statistics for Pedestrian Detection: Combining Appearance, Motion and Stereo
32 Citations2010Stefan Walk, Konrad Schindler +1 more
This paper proposes a simple yet highly effective novel feature based on binocular disparity, outperforming previ-ously proposed stereo features and shows that the combination of different classifiers often improves performance even when classifiers are based on the same feature or feature combination.
Lecture notes in computer scienceHigh-Level Fusion of Depth and Intensity for Pedestrian Classification
28 Citations2009Marcus Rohrbach, Markus Enzweiler +1 more
A novel approach to pedestrian classification which involves a high-level fusion of depth and intensity cues, which demonstrates a significant performance improvement of the combined intensity-depth representation over depth-only and intensity-only models.