Filters








1,080 Hits in 3.3 sec

A Robust Background Initialization Algorithm with Superpixel Motion Detection [article]

Zhe Xu, Biao Min, Ray C.C. Cheung
2018 arXiv   pre-print
In this paper, a robust background initialization approach based on superpixel motion detection is proposed.  ...  A subsequence with stable illumination condition is first selected for background estimation.  ...  In this paper, a robust background initialization algorithm based on superpixel motion detection with relatively low computational complexity is proposed.  ... 
arXiv:1805.06737v1 fatcat:7q6rszwaczgaxktjvsrgk3pq5e

Co-occurrence Background Model with Superpixels for Robust Background Initialization [article]

Wenjun Zhou, Yuheng Deng, Bo Peng, Dong Liang, Shun'ichi Kaneko
2020 arXiv   pre-print
such as illumination changes,background motion,camera jitter and intermittent movement,etc.In this paper,we develop a co-occurrence background model with superpixel segmentation for robust background  ...  We first introduce a novel co-occurrence background modeling method called as Co-occurrence Pixel-Block Pairs(CPB)to generate a reliable initial background model,and the superpixel segmentation is utilized  ...  CONCLUSIONS In this paper, we propose a new approach for robust background initialization of a complex scene based on cooccurrence background model (CPB) with superpixel segmentation.  ... 
arXiv:2003.12931v1 fatcat:najugnc445h7pjulzntekte5eu

An Adaptive Superpixel Based Hand Gesture Tracking and Recognition System

Hong-Min Zhu, Chi-Man Pun
2014 The Scientific World Journal  
Finally, the hand gesture is recognized by the extracted motion trajectory with a trained SVM classifier.  ...  First we employed the motion detection of superpixels and unsupervised image segmentation to detect the moving target hand using the first few frames of the input video sequence.  ...  The initial hand detection and model construction procedure is summarized in Algorithm 1.  ... 
doi:10.1155/2014/849069 pmid:24991650 pmcid:PMC4058504 fatcat:lvrjdwfkr5feff6sgzdgc2lsca

Motion segmentation with occlusions on the superpixel graph

Alper Ayvaci, Stefano Soatto
2009 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops  
It relies on a finer partitioning of the image domain into regions of uniform photometric properties, with motion segments made of unions of such "superpixels."  ...  We present a motion segmentation algorithm that partitions the image plane into disjoint regions based on their parametric motion.  ...  However, partially occluded superpixels still remain as a problem which will be solved with the proposed occlusion detection algorithm.  ... 
doi:10.1109/iccvw.2009.5457630 dblp:conf/iccvw/AyvaciS09 fatcat:ruero2eddndtvhcxj356fnjnau

An Adaptive Superpixel Tracker Using Multiple Features

Jingjing Liu, Bin Zhang, Xu Cheng, Ying Chen, Li Zhao
2019 Computers Materials & Continua  
In this paper, a superpixel-based adaptive tracking algorithm by using color histogram and haar-like feature is proposed, whose feature is classified into the middle level.  ...  of background subtraction method as the frame feature.  ...  Figure 3 : 3 The accuracy-robustness ranking plots with respect to the two experiments 1. Algorithm 1. Main steps of the proposed tracking algorithm Initialization: For t=1 to m (e.g., m=4) 1.  ... 
doi:10.32604/cmc.2019.05968 fatcat:f4k62lyoqzawvjprprykvjv6hi

Superpixel-based video object segmentation using perceptual organization and location prior

Daniela Giordano, Francesca Murabito, Simone Palazzo, Concetto Spampinato
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Initially, we rapidly generate a coarse foreground segmentation, which provides predictions about motion regions by analyzing how superpixel segmentation changes in consecutive frames.  ...  We then exploit these location priors to refine the initial segmentation by optimizing an energy function based on appearance and perceptual organization, only on regions where motion is observed.  ...  The last evaluation was carried out on the I2R Dataset [10] , which contains nine videos (at 120×160 resolution) taken with static cameras showing people in different indoor and outdoor scenes.  ... 
doi:10.1109/cvpr.2015.7299114 dblp:conf/cvpr/GiordanoMPS15 fatcat:b5lwu4tdqbfe5ggmofog7zpkrm

Unsupervised Single Moving Object Detection Based on Coarse-to-Fine Segmentation

2016 KSII Transactions on Internet and Information Systems  
Given the sparsely labelled trajectory points, we adopt a coarse-to-fine strategy to detect and segment the foreground from the background.  ...  According to the experiments, our algorithm is effective and efficient, and favorable results can be achieved compared with state-of-the-art methods.  ...  In contrast, the latter kind of algorithms detect moving objects by using different types of motion cues. A kind of widely used motion cues is optical flow computed between adjacent frames.  ... 
doi:10.3837/tiis.2016.06.012 fatcat:46icxv222zb4th4d2zi5wtdxku

Robust Video Object Segmentation Via Propagating Seams and Matching Superpixels

Yun Liang, Yuqing Zhang, Yihan Wu, Shuqin Tu, Caixing Liu
2020 IEEE Access  
This paper proposes a robust video object segmentation method by propagating patch seams and matching superpixels.  ...  Video object segmentation aims at separating foreground object from background, and it is far from well solved for different challenges such as deformation, occlusion and motion blurs.  ...  , fast motion, cluttered background, and so on.  ... 
doi:10.1109/access.2020.2981140 fatcat:dunhm247trddbnjtfhynfyjxfu

AUTOMATIC IDENTIFICATION OF ANIMAL USING VISUAL AND MOTION SALIENCY

G.N.Murugananthan .
2014 International Journal of Research in Engineering and Technology  
., not limited to any particular type of object).To separate foreground and background regions within and across video frames, the proposed method utilizes visual and motion saliency information extracted  ...  A conditional random field is applied to effectively combine the saliency induced features, which allows us to deal with unknown pose and scale variations of the foreground object (and its articulated  ...  He started with the watershed segmentation algorithm to provide initial regions as per the pixels luminance gradient, then regions are merged according to their color and motion connection.  ... 
doi:10.15623/ijret.2014.0304012 fatcat:3gwgnlw75rdi7pzepwjsbqztn4

Foreground Segmentation via Dynamic Tree-Structured Sparse RPCA [chapter]

Salehe Erfanian Ebadi, Ebroul Izquierdo
2016 Lecture Notes in Computer Science  
Video analysis often begins with background subtraction which consists of creation of a background model, followed by a regularization scheme.  ...  using our approximated Robust Principal Component Analysis method extended to handle camera motion.  ...  Acknowledgments The research leading to this paper was fully supported by the LASIE project (http://www.lasie-project.eu/) with funding from the European Unions Seventh Framework Program for research and  ... 
doi:10.1007/978-3-319-46448-0_19 fatcat:hmpxtn5k7bcvrpwqkqvgjxj57i

Improving the Efficiency of Background Subtraction using Superpixel Extraction and Midpoint for Centroid

K. SuganyaDevi, N. Malmurugan, S. Poornima
2012 International Journal of Computer Applications  
This paper deals with an efficient background subtraction of image/frames of video by improving the execution speed, accuracy and reduce the usage of memory.  ...  Three important techniques are applied to improve the efficiency: superpixel extraction, canny edge detection and fuzzy c means.  ...  Initialize the first pixel value P 1 as initial value of the super pixel 2. Get a pixel Pi 3. Compare Pi with P 1 i.e compute P i -P 1 4.  ... 
doi:10.5120/6136-8372 fatcat:uqgcftv2ovdgbjhbbz46io5qia

Coarse-to-fine online learning for hand segmentation in egocentric video

Ying Zhao, Zhiwei Luo, Changqin Quan
2018 EURASIP Journal on Image and Video Processing  
Zhiwei Luo is a professor and Changqin Quan is an associate professor in Graduate School of System Informatics at Kobe University, Kobe, Japan.  ...  Authors' contributions YZ generated this idea, implemented the core algorithm, designed all the experiments, addressed the resulting data, and drafted the manuscript.  ...  After detecting the presence of hands, we initially segment moving hand regions based on motion and appearance clustering.  ... 
doi:10.1186/s13640-018-0262-1 fatcat:4nzgfsommngi3c2kopspmuzoe4

Object Tracking using Superpixel Confidence Map in Centroid Shifting Method

Richard Evan Sutanto, , Lenny, Suk Ho Lee
2016 Indian Journal of Science and Technology  
The experiment will compare proposed method with other previous algorithms, original tracking based on centroid shifting and super pixel tracking using a same dataset.  ...  Before it used super pixel confidence map value, it will do a centroid shifting based to find target location by weighted the area with mean of the centroids comparing to each color bin of the target.  ...  In 9 and 10 also has proposed a method based on robust tracker to do motion tracking using super pixel algorithm, they use robust tracker based on a discriminative appearance model and super pixel.  ... 
doi:10.17485/ijst/2016/v9i35/101783 fatcat:mdvllhas5jgijlmenoskfaccfm

Superpixel tracking

Shu Wang, Huchuan Lu, Fan Yang, Ming-Hsuan Yang
2011 2011 International Conference on Computer Vision  
While numerous algorithms have been proposed for object tracking with demonstrated success, it remains a challenging problem for a tracker to handle large change in scale, motion, shape deformation with  ...  We present a discriminative appearance model based on superpixels, thereby facilitating a tracker to distinguish the target and the background with mid-level cues.  ...  Table 1 . 1 Proposed algorithm. Initialization: for t = 1 to m (e.g., m is set to 4 in all experiments) 1. Initialize parameters of our algorithm in the first frame. 2.  ... 
doi:10.1109/iccv.2011.6126385 dblp:conf/iccv/WangLYY11 fatcat:clpwfhbxjrfr7iax5nxfwphaoe

Unsupervised Video Object Segmentation using Motion Saliency-Guided Spatio-Temporal Propagation [article]

Yuan-Ting Hu, Jia-Bin Huang, Alexander G. Schwing
2018 arXiv   pre-print
Our approach leads to significantly better initial foreground-background estimates and their robust as well as accurate diffusion across time.  ...  However, to date, fast motion, motion blur and occlusions pose significant challenges.  ...  Intuitively, boundary pixels largely correspond to background and pixels with a similar motion are likely background too.  ... 
arXiv:1809.01125v1 fatcat:d6vwfn6ypfa5rd2b7gpggks5k4
« Previous Showing results 1 — 15 out of 1,080 results