Filters








1,003 Hits in 6.3 sec

From Superpixel to Human Shape Modelling for Carried Object Detection [article]

Farnoosh Ghadiri, Robert Bergevin, Guillaume-Alexandre Bilodeau
2018 arXiv   pre-print
Detecting carried objects is one of the requirements for developing systems to reason about activities involving people and objects.  ...  We present an approach to detect carried objects from a single video frame with a novel method that incorporates features from multiple scales.  ...  In our previous work (ECE) [6] , we introduced a carried object detector based on an ensemble of contour exemplars.  ... 
arXiv:1801.03551v1 fatcat:dtcb2e75png2rj2bckbe6q7dy4

Semi-supervised Affinity Propagation Clustering Algorithm Based on Stratified Combination

Zhen Zhang, Bin-qiang Wang, Peng Yi, Ju-long Lan
2014 Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology  
The experimental results on the Wisconsin Diagnostic Breast Cancer (WDBC) datasets show a higher predictive accuracy.  ...  In the first phase, the Affinity Propagation (AP) clustering method is used as an instance reduction technique, which can find noisy instances and eliminate them.  ...  Although both reached a 98.45% accuracy, PSO-KDE is better than GA-KDE because of less selected features. Ensemble feature selection based on bi-objective genetic algorithm can be found in [32] .  ... 
doi:10.3724/sp.j.1146.2012.00673 fatcat:kelpr4jhengvnfyz7xhyj72g3y

Fully Automated Integrated Segmentation of Carotid Artery Ultrasound Images Using DBSCAN and Affinity Propagation

S. Latha, Dhanalakshmi Samiappan, P. Muthu, R. Kumar
2021 Journal of Medical and Biological Engineering  
Methods A fully automatic self-learning based segmentation is proposed by extracting the edges by a modified affinity propagation, which are given as inputs to the Density Based Spatial Clustering of Applications  ...  The proposed approach gives an improved accuracy of 12% increase when compared with the manual segmentation and 15% compared with segmentation by affinity propagation and DBSCAN when performed individually  ...  Funding The work is funded by The Institution of Engineers (India). Data Availability Data available on request from the authors. Code Availability Code available on request from the authors.  ... 
doi:10.1007/s40846-020-00586-9 fatcat:rxxwsryxxnehnouqqqmyvhjd4q

Automatic facial landmark labeling with minimal supervision

Yan Tong, Xiaoming Liu, Frederick W. Wheeler, Peter Tu
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
Landmark labeling of training images is essential for many learning tasks in computer vision, such as object detection, tracking, and alignment.  ...  Our approach, named semi-supervised least-squares congealing, aims to minimize an objective function defined on both labeled and unlabeled images.  ...  This paper focuses on geometric/landmark knowledge labeling, which is typically carried out manually. Practical applications, such as object detection, often require thou- Figure 1 .  ... 
doi:10.1109/cvpr.2009.5206670 dblp:conf/cvpr/TongLWT09 fatcat:rvafj4xcbjdzjohte3qjgjrfpe

Automatic facial landmark labeling with minimal supervision

Yan Tong, Xiaoming Liu, F.W. Wheeler, P. Tu
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
Landmark labeling of training images is essential for many learning tasks in computer vision, such as object detection, tracking, and alignment.  ...  Our approach, named semi-supervised least-squares congealing, aims to minimize an objective function defined on both labeled and unlabeled images.  ...  This paper focuses on geometric/landmark knowledge labeling, which is typically carried out manually. Practical applications, such as object detection, often require thou- Figure 1 .  ... 
doi:10.1109/cvprw.2009.5206670 fatcat:vslpuveabjbcrbrpq4cecklzjq

Automatic Detection and Classification of Breast Tumors in Ultrasonic Images Using Texture and Morphological Features

Yanni Su
2011 Open Medical Informatics Journal  
By incorporating local features of texture and position, a ROI is firstly detected using a self-organizing map neural network.  ...  Due to severe presence of speckle noise, poor image contrast and irregular lesion shape, it is challenging to build a fully automatic detection and classification system for breast ultrasonic images.  ...  defined based on physician's experience.  ... 
doi:10.2174/1874431101105010026 pmid:21892371 pmcid:PMC3158436 fatcat:2vwngkay3vg4hn5corwhfama6q

Multiscale Object Detection from Drone Imagery Using Ensemble Transfer Learning

Rahee Walambe, Aboli Marathe, Ketan Kotecha
2021 Drones  
In this paper, we present an implementation of ensemble transfer learning to enhance the performance of the base models for multiscale object detection in drone imagery.  ...  Our approach is more practical and efficient due to the use of transfer learning and two-level voting strategy ensemble instead of training custom models on entire datasets.  ...  In 2011, an ensemble Exemplar SVM framework [22] achieved comparable SOTA performance to the complex latent part-based model of Felzenszwalb et al. [24] .  ... 
doi:10.3390/drones5030066 fatcat:pikxnef2dfhmndro35br2rz4vi

The Impact of Density and Ratio on Object-Ensemble Representation in Human Anterior-Medial Ventral Visual Cortex

Jonathan S. Cant, Yaoda Xu
2014 Cerebral Cortex  
In Experiment 2, we varied relative density by changing the ratio of 2 types of objects comprising an ensemble, and observed significant sensitivity in PPA to such ratio change.  ...  Object-ensemble processing in this region may thus depend on high-level visual information, such as object ratio, rather than low-level information, such as spacing/spatial frequency.  ...  Notes Conflict of Interest: None declared.  ... 
doi:10.1093/cercor/bhu145 pmid:24964917 pmcid:PMC4626831 fatcat:kmyuhxtyibhdpl242zgtujfj6e

Detecting animals in African Savanna with UAVs and the crowds

Nicolas Rey, Michele Volpi, Stéphane Joost, Devis Tuia
2017 Remote Sensing of Environment  
It relies on an animal-detection system based on machine learning, trained with crowd-sourced annotations provided by volunteers who manually interpreted sub-decimeter resolution color images.  ...  It shows that the detection of large mammals in semi-arid Savanna can be approached by processing data provided by standard RGB cameras mounted on affordable fixed wings UAVs.  ...  Friedrich Reinhard of Kuzikus Wildlife Reserve, Namibia) and the QCRI and Micromappers (in particular Dr. Ferda Ofli and Ji Kim Lucas) for the support in the collection of ground truth data.  ... 
doi:10.1016/j.rse.2017.08.026 fatcat:w7whh5zeofdzxezqbe74ae5vvy

Combining Stacked Denoising Autoencoders and Random Forests for Face Detection [chapter]

Jingjing Deng, Xianghua Xie, Michael Edwards
2016 Lecture Notes in Computer Science  
Notably, Deep Neural Network (DNN) based methods have been found to outperform most traditional detectors in a multitude of studies, employing deep network structures and complex training procedures.  ...  Detecting faces in the wild is a challenging problem due to large visual variations introduced by uncontrolled facial expressions, head pose, illumination and so on.  ...  Meanwhile, detection speed will suffer due to exploring larger search space. Contour based object tracking method such as [5] can also be applied to face detection task.  ... 
doi:10.1007/978-3-319-48680-2_31 fatcat:e2ffauzdabbs5kwy4ldy3lt3y4

Integrating contour and skeleton for shape classification

Xiang Bai, Wenyu Liu, Zhuowen Tu
2009 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops  
derive an effective classifier. (2) We collect a challenging shape database in which there are 20 categories of animals, with each having 100 shapes.  ...  A thorough experimental study is conducted showing significant improvement by the proposed algorithm over many of the state-of-the-art shape matching and classification algorithms, on both our dataset  ...  Any findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the Office of Naval Research.  ... 
doi:10.1109/iccvw.2009.5457679 dblp:conf/iccvw/BaiLT09 fatcat:b3p6y3uyhjgxbd5sdwhm27wixu

Robust Visual Tracking via Collaborative and Reinforced Convolutional Feature Learning

Dongdong Li, Yangliu Kuai, Gongjian Wen, Li Liu
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
In the paper, we design an end-to-end trainable tracking framework based on Siamese network, which proposes to learn the low-level fine-grained and high-level semantic representations simultaneously with  ...  Extensive experimental tracking results on the widely used OTB and TC128 benchmarks demonstrate the superiority of our tracker. Meanwhile, our proposed tracker can achieve a real-time tracking speed.  ...  The two networks are used interchangeably based on distractor detection scheme. Chao et al.  ... 
doi:10.1109/cvprw.2019.00085 dblp:conf/cvpr/LiKWL19 fatcat:mwzpnvf7w5fjxflrnoxcqnp4h4

Automatic Niching Differential Evolution with Contour Prediction Approach for Multimodal Optimization Problems

Zi-Jia Wang, Zhi-Hui Zhan, Ying Lin, Wei-Jie Yu, Hua Wang, Sam Kwong, Jun Zhang
2019 IEEE Transactions on Evolutionary Computation  
In this paper, we propose a new automatic niching technique based on the affinity propagation clustering (APC) and design a novel niching differential evolution (DE) algorithm, termed as automatic niching  ...  However, most of the existing niching techniques are either sensitive to the niching parameters or require extra fitness evaluations (FEs) to maintain the niche detection accuracy.  ...  For example, pedestrian detection often requires to extract multiple pedestrian from a given image [3] . Locating all the global optima of an MMOP has significant benefits.  ... 
doi:10.1109/tevc.2019.2910721 fatcat:436wfoihpvawtc6mmv6b5wl5sa

Human activities recognition using depth images

Raj Gupta, Alex Yong-Sang Chia, Deepu Rajan
2013 Proceedings of the 21st ACM international conference on Multimedia - MM '13  
images to recognize an activity.  ...  We exploit unique poses of an activity and the temporal ordering of these poses to learn subsequences of codewords which are strongly discriminative for the activity.  ...  In the 'personseen-before' setting, we evaluate the boosted ensemble on video sequences for which the ensemble has previously seen the person carrying out the activity.  ... 
doi:10.1145/2502081.2502099 dblp:conf/mm/GuptaCR13 fatcat:bugd4sai5fggzmyn3k7fm6ooja

Measuring visual form discrimination with blur thresholds

G. Westheimer
2013 Journal of Vision  
But there are ways of carrying out a stress test on form discrimination other than making the tokens smaller.  ...  The differences are rather in the decision criteria on which the responses are based.  ... 
doi:10.1167/13.5.13 pmid:23599417 fatcat:eoggy5yfa5bnfifzk35vntbqcu
« Previous Showing results 1 — 15 out of 1,003 results