Filters








3,722 Hits in 3.6 sec

Learning pedestrian models for silhouette refinement

Lee, Dalley, Tieu
2003 Proceedings Ninth IEEE International Conference on Computer Vision  
We apply our silhouette extraction approach to the NIST gait data set and show that under the gait recognition task, our model-based sulhouettes result in much higher recognition rates than silhouettes  ...  We present a model-based method for accurate extraction of pedestrian silhouettes from video sequences.  ...  Our silhouette models can be used in two ways: to fill in silhouettes for any algorithm that needs accurate silhouette sequences, and to be used directly for gait recognition.  ... 
doi:10.1109/iccv.2003.1238411 dblp:conf/iccv/LeeDT03 fatcat:amakmut3cvc7xe5szy3ethzoai

Boosting Shape Classifiers Accuracy by Considering the Inverse Shape

Sébastien Piérard, Antoine Lejeune, Marc Van Droogenbroeck
2016 Journal of Pattern Recognition Research  
This opens the path for better techniques that could use, as a rule of thumb, both the representations of a shape and its inverse for the purpose of classification.  ...  Our experiments establish that, for shape classification, a representation integrating the inverse shape often outperforms a representation restricted to the shape.  ...  silhouettes filled Figure 9 9 Figure 9: Classification accuracy obtained for human silhouette recognition based on Barnich' method [4], with respect to the proportion of rectangles taken from the shape  ... 
doi:10.13176/11.727 fatcat:miwz3b4ijjgi3ahb4hipcod6ce

Frame difference energy image for gait recognition with incomplete silhouettes

Changhong Chen, Jimin Liang, Heng Zhao, Haihong Hu, Jie Tian
2009 Pattern Recognition Letters  
The frieze and wavelet features are adopted and hidden Markov model (HMM) is employed for recognition.  ...  The quality of human silhouettes has a direct effect on gait recognition performance.  ...  GEI performs much better than GHI and GMI, but little worse than IDTW. Wavelet(FD) has the best recognition rates for most angles and reaches 100% for the first three angles.  ... 
doi:10.1016/j.patrec.2009.04.012 fatcat:xydmg34egvbtpeni3ikix37jd4

Human Motion Analysis by Fusion of Silhouette Orientation and Shape Features

D.K. Vishwakarma, Ashish Dhiman, Rockey Maheshwari, Rajiv Kapoor
2015 Procedia Computer Science  
The proposed method provides the advantage of merging the local and global features of the silhouette and thus provides the discriminative feature representation for human activity recognition.  ...  This paper presents a simple and competent approach for Human Activity Recognition based on the hypothesis that every action/activity has some kind of rotation information with translation.  ...  Ahmad et al. 19 , formed the silhouette energy images, to differentiate the shape and motion properties for human actions recognition.  ... 
doi:10.1016/j.procs.2015.07.515 fatcat:pgs2k5aquvd6ha3g63q2wuj47q

A New Automatic Gait Recognition method based on the Perceptual Curve

Yanmei Chai, Qing Wang, Rongchun Zhao, Changzhu Wu
2005 TENCON 2005 - 2005 IEEE Region 10 Conference  
Then we use inner boundary tracking algorithm to extract the walker's binary silhouettes and compute the PSD (Perceptual Shape Descriptor) of the shapes.  ...  In this paper we proposed a new automatic gait recognition approach based on the perceptual curve, which can effectively preserve the temporal changes of the walker's silhouette shape consistency with  ...  ACKNOWLEDGMENT The work described in the paper was supported by Aerospace Innovation Fund, Aviation Science Fund (03I53065) and National Natural Science Fund (60403008), P. R. China.  ... 
doi:10.1109/tencon.2005.300859 fatcat:5uubxb6w2jedvi75pg6nvhgm74

Effect of Silhouette Quality on Hard Problems in Gait Recognition

Z. Liu, S. Sarkar
2005 IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics)  
that the observed poor performance of gait recognition for hard problems involving matching across factors such as surface, time, and shoe are not due to poor silhouette quality, beyond what is available  ...  an eigen-stance model, to correct for common errors in silhouette detection arising from shadows and background subtraction.  ...  Malave for running some of the early experiments with manual silhouettes, the gait researchers at CMU, MIT, and Southampton for helping to correct some of the errors in the manual silhouettes, and the  ... 
doi:10.1109/tsmcb.2004.842251 pmid:15828648 fatcat:lrcph2kgvjdsbbhfxd3syloxv4

Motion segmentation and pose recognition with motion history gradients

Gary R. Bradski, James W. Davis
2002 Machine Vision and Applications  
This paper uses a simple method for representing motion in successively layered silhouettes that directly encode system time termed the timed Motion History Image (tMHI).  ...  This representation can be used to both (a) determine the current pose of the object and to (b) segment and measure the motions induced by the object in a video scene.  ...  Mahalanobis Match to Hu Moments of Silhouette Pose For recognition of silhouette pose, seven higher-order Hu moments [19] provide shape descriptors that are invariant to translation and scale.  ... 
doi:10.1007/s001380100064 fatcat:jerta6d5gzhuxasb2rs3ciao2a

Robust arbitrary view gait recognition based on parametric 3D human body reconstruction and virtual posture synthesis

Jian Luo, Jin Tang, Tardi Tjahjadi, Xiaoming Xiao
2016 Pattern Recognition  
The GEI reduces storage space and computation time, 52 and can achieve high recognition rate when the gait cycle is complete.  ...  By constructing an over-complete dictionary of 3D gait and simultaneous sparse 63 representation model, the simultaneous sparse representation vectors are reconstructed for 3D probe gait models 64 which  ...  Rank-1 recognition rates of AVGR-BPRS, GEIs and GFI for walking normally in 90° view with random number recognition with random number of probe frames 629 Fig. 18 shows the rank-1 recognition rates of  ... 
doi:10.1016/j.patcog.2016.05.030 fatcat:szs7qz3yqbfalnsrns6spje57y

Automatic gait recognition based on statistical shape analysis

Liang Wang, Tieniu Tan, Weiming Hu, Huazhong Ning
2003 IEEE Transactions on Image Processing  
This paper aims to propose a simple and efficient automatic gait recognition algorithm using statistical shape analysis.  ...  Supervised pattern classification techniques based on the full Procrustes distance measure are adopted for recognition.  ...  Our thanks also go to those who have contributed to the establishment of the NLPR database, and to the anonymous referees for their thorough review and constructive comments.  ... 
doi:10.1109/tip.2003.815251 pmid:18237983 fatcat:5eahkomnojcnxlq53fbrjl6xii

ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness [article]

Robert Geirhos, Patricia Rubisch, Claudio Michaelis, Matthias Bethge, Felix A. Wichmann, Wieland Brendel
2019 arXiv   pre-print
This provides a much better fit for human behavioural performance in our well-controlled psychophysical lab setting (nine experiments totalling 48,560 psychophysical trials across 97 observers) and comes  ...  Convolutional Neural Networks (CNNs) are commonly thought to recognise objects by learning increasingly complex representations of object shapes.  ...  for support and/or useful discussions.  ... 
arXiv:1811.12231v2 fatcat:63xbjleue5gmbattknf5skegum

Autonomous Multiple Gesture Recognition System for Disabled People

Amarjot Singh, John Buonassisi, Sukriti Jain
2014 International Journal of Image Graphics and Signal Processing  
Pose is extracted on the basis of silhouettes using timed motion history (tMHI) followed by gesture recognition with Hu-Moments. Signs involving motion are recognized with the help of optical flow.  ...  The system makes use of pose estimation for 10 signs used by hearing impaired people to communicate.  ...  The advantage with this algorithm is that it makes our representation independent of the speed of the system or the frame rate such that irrespective of the capture rates, the same MHI area will be covered  ... 
doi:10.5815/ijigsp.2014.02.05 fatcat:abihvg7r5ncxji3tlkypezodvm

Vision based Human Gait Recognition System: Observations, Pragmatic Conditions and Datasets

V. G. Manjunatha Guru, V. N. Kamalesh
2015 Indian Journal of Science and Technology  
Findings: This paper provides a comprehensive survey of current developments on gait recognition approaches and emphasizes on three major phases involved in gait recognition system, namely representation  ...  For this reason, human gait recognition at a distance recently gained wider interest from research community.  ...  Wang et al. 8 apply the procreates shape analysis to silhouette shapes and extract a procreates mean shape from a sequence of silhouettes as gait signature. Then it is followed by PCA.  ... 
doi:10.17485/ijst/2015/v8i15/71237 fatcat:nnwaj2ukl5gd3lzmxith4qh3nu

View Invariant Gait Recognition [chapter]

Richard D. Seely, Michela Goffredo, John N. Carter, Mark S. Nixon
2009 Advances in Pattern Recognition  
Gait as a biometric has now shown increasing recognition capability. There are many approaches and these show that recognition can achieve excellent performance on large databases.  ...  These methods use several video cameras and the 3D reconstruction is achieved after a camera calibration process.  ...  [16] introduced a new representation for human gait recognition, called mass vector and defined as the number of pixels with a nonzero value in a given row of the silhouette.  ... 
doi:10.1007/978-1-84882-385-3_3 fatcat:f5kmsdw3yzfh3emuyok7dj2ca4

Multi-view human action recognition using 2D motion templates based on MHIs and their HOG description

Fiza Murtaza, Muhammad Haroon Yousaf, Sergio A. Velastin
2016 IET Computer Vision  
, 2) requires less memory and bandwidth requirements and hence 3) is computationally efficient which makes it suitable for real-time action recognition.  ...  The proposed method is evaluated on the new MuHAVi-uncut dataset having a large number of action categories and a large set of camera-views with noisy silhouettes.  ...  Shape based methods do (a) (b) not depend upon moving person's clothes color, texture, and luminance therefore these are potentially useful for action representation.  ... 
doi:10.1049/iet-cvi.2015.0416 fatcat:mz7kzndxrbdtfh3xahbupe3n3u

Dense RGB-D Map-Based Human Tracking and Activity Recognition using Skin Joints Features and Self-Organizing Map

2015 KSII Transactions on Internet and Information Systems  
During human tracking and activity recognition, initially, dense depth images are captured using depth camera.  ...  In order to track human silhouettes, we considered spatial/temporal continuity, constraints of human motion information and compute centroids of each activity based on chain coding mechanism and centroids  ...  However, binary silhouette itself has very limited information due to flat pixel intensity value's (i.e. 0 and 1) which causes low recognition rate especially in case of complex poses and self-occlusions  ... 
doi:10.3837/tiis.2015.05.017 fatcat:lopculvkwfcfvocamxz4ty6aqe
« Previous Showing results 1 — 15 out of 3,722 results