759 Hits in 10.5 sec

Research on Salient Object Detection using Deep Learning and Segmentation Methods

2019 International journal of recent technology and engineering  
Detecting and segmenting salient objects in natural scenes, often referred to as salient object detection has attracted a lot of interest in computer vision and recently various heuristic computational  ...  The aim of this review work is to study about the details of methods in salient object detection.  ...  Eye trackers are expensive and interactive techniques for a more memory consumption. 4 Salient object detection via color and texture cues Zhang, [2017] Bottom-up salient object detection approach  ... 
doi:10.35940/ijrte.b1046.0982s1119 fatcat:6ofq53vb7zhx7boq4ndpraphs4

Robust 3D Action Recognition through Sampling Local Appearances and Global Distributions [article]

Mengyuan Liu, Hong Liu, Chen Chen
2017 arXiv   pre-print
Our model performs favorably compared with common STIP detection and description methods.  ...  Using the Bag-of-Visual-Words (BoVW) model, motion and shape cues are combined to form a fused action representation.  ...  Second, previous methods on depth action recognition define local points with salient motion as STIPs, while local points with distinctive shape cues are not detected or are ignored.  ... 
arXiv:1712.01090v2 fatcat:pwfadl3qq5cudnxhr5ek24gjlq

What Can We Learn from Biological Vision Studies for Human Motion Segmentation? [chapter]

Cheng Chen, Guoliang Fan
2006 Lecture Notes in Computer Science  
Specifically, we discuss the roles and interactions of bottom-up and top-down processes in visual perception processing as well as how to combine them synergistically in one computational model to guide  ...  According to this model, object segmentation, motion estimation, and action recognition are results of recurrent feedforward (bottom-up) and feedback (top-down) processes.  ...  In other words, human vision recognizes salient objects in space and time simultaneously.  ... 
doi:10.1007/11919629_79 fatcat:2uark6vbibdbjosbrz6p3pqpxi

A Review on Computer Vision-Based Methods for Human Action Recognition

Mahmoud Al-Faris, John Chiverton, David Ndzi, Ahmed Isam Ahmed
2020 Journal of Imaging  
However, accurate and effective vision based recognition systems continue to be a big challenging area of research in the field of computer vision.  ...  To this end, the direction of this research is sorted out from hand-crafted representation based methods including holistic and local representation methods with various sources of data, to a deep learning  ...  In general, there has been much success with 2D and 3D CNN in e.g., image classification, object recognition, speech recognition and action recognition.  ... 
doi:10.3390/jimaging6060046 pmid:34460592 pmcid:PMC8321068 fatcat:eyp2pu6egzcunagferl7dhffay

Vision, Attention Control, and Goals Creation System [chapter]

Konstantinos Rapantzikos, Yannis Avrithis, Stefanos Kolias
2010 Perception-Action Cycle  
Rapantzikos ( ) Image, Video and  ...  Hence a second question arises related to how these channels are "linked" in order to provide useful information to the brain.  ...  Recently, points of interest combined with bag-of-words approaches have been also used for object/event detection and recognition.  ... 
doi:10.1007/978-1-4419-1452-1_11 fatcat:cyydaayk3zauncho44knonuqfm

Developmental approach for interactive object discovery

Natalia Lyubova, David Filliat
2012 The 2012 International Joint Conference on Neural Networks (IJCNN)  
A hierarchical object representation is constructed from SURF points and color of superpixels that are grouped in local geometric structures and form the basis of a multiple-view object model.  ...  The learning algorithm accumulates the statistics of feature occurrences and identifies objects using a maximum likelihood approach and temporal coherency.  ...  In order to process image faster and more efficiently than operating on pixels intensities, local descriptors are often used to characterize stable image patches at salient positions.  ... 
doi:10.1109/ijcnn.2012.6252606 dblp:conf/ijcnn/LyubovaF12 fatcat:mleavkf4w5dtzgi7e6vw4jmzy4

Visual Saliency Detection Using Group Lasso Regularization in Videos of Natural Scenes

Nasim Souly, Mubarak Shah
2015 International Journal of Computer Vision  
Thus, its applications for complex tasks such as object detection, object recognition and video compression have attained interest in computer vision studies.  ...  In addition, we show our video saliency can be used to improve Communicated by the performance of human action recognition on a standard dataset.  ...  Local spatio-temporal descriptors are being widely used for action recognition in videos.  ... 
doi:10.1007/s11263-015-0853-6 fatcat:2sdtk73gtnaw5hou2aedvmxgja

Modeling attention to salient proto-objects

Dirk Walther, Christof Koch
2006 Neural Networks  
We demonstrate that the suggested model can enable a model of object recognition in cortex to expand from recognizing individual objects in isolation to sequentially recognizing all objects in a more complex  ...  We propose a biologically plausible model of forming and attending to proto-objects in natural scenes.  ...  Acknowledgements Parts of the work presented in this paper originated from collaborations with Laurent Itti, Maximilian Riesenhuber, Tomaso Poggio, Ueli Rutishauser, and Pietro Perona.  ... 
doi:10.1016/j.neunet.2006.10.001 pmid:17098563 fatcat:idkut7zu3jgnxb6qtzfqvmkygq

Human Activity Recognition Using Cascaded Dual Attention CNN and Bi-Directional GRU Framework [article]

Hayat Ullah, Arslan Munir
2022 arXiv   pre-print
The extracted discriminative salient features are then forwarded to stacked bi-directional gated recurrent unit (Bi-GRU) for long-term temporal modeling and recognition of human actions using both forward  ...  in their proposals to deal with challenging human activity recognition problem.  ...  attention mechanism focuses on object saliency in the given feature maps by paying more attention to important features across each color channel and localizing salient regions.  ... 
arXiv:2208.05034v1 fatcat:zxzb7rn32vajzkbwngu3ajztdq

Adaptive scene dependent filters for segmentation and online learning of visual objects

J.J. Steil, M. Götting, H. Wersing, E. Körner, H. Ritter
2007 Neurocomputing  
We apply the ASDF hierarchy for preprocessing input images in a feature-based biologically motivated object recognition learning architecture. and show experiments with this real-time vision system running  ...  It is based on forming a combined feature space from basic feature maps like, color, disparity, and pixel position.  ...  We use this to subtract regions representing skin and hand color that are detected using a separate specialized skin color detection.  ... 
doi:10.1016/j.neucom.2006.11.020 fatcat:4trbdjt3rfcytdbop25o35gzne

An active vision architecture based on iconic representations

Rajesh P.N. Rao, Dana H. Ballard
1995 Artificial Intelligence  
These representations are stored in two separate memories. One memory is indexed by image coordinates while the other is indexed by object coordinates.  ...  Object identification matches a fovea1 set of image features with all possible model features.  ...  of the images used in the experiments, Garbis Salgian for implementing some of the first routines on the Datacube MV200 system, Jim Vallino for discussions regarding the MV200, and Lambert Wixson for  ... 
doi:10.1016/0004-3702(95)00026-7 fatcat:csnwov5enneilpsgahnelet4hu

Cluttered TextSpotter: An End-to-End Trainable Light- weight Scene Text Spotter for Cluttered Environment

Randheer Bagi, Tanima Dutta, Hari Prabhat Gupta
2020 IEEE Access  
It helps to localize in scene images with background clutters, where partially occluded text parts, truncation artifacts, and perspective distortions are present.  ...  Scene text detection and recognition approaches have received immense attention in computer vision research community.  ...  In generic object detection using deep neural networks (in short deep networks), global structures and context information of an object are used to learn and discriminate salient features [9] .  ... 
doi:10.1109/access.2020.3002808 fatcat:x4kbcajahrc5vgtuxsc6oyyjsa

Real-time Color Ball Tracking for Augmented Reality [article]

Daniel Sýkora, David Sedlácek, Kai Riege
2008 ICAT-EGVE 2014 - International Conference on Artificial Reality and Telexistence and Eurographics Symposium on Virtual Environments  
It is fast enough to be easily combined with another real-time tracking engine.  ...  In this paper, we introduce a light-weight and robust tracking technique based on color balls.  ...  Another problem arises when non-circular objects with similar colors appear in the image.  ... 
doi:10.2312/egve/egve08/009-016 fatcat:skv6lo5sjjcnfhas3ilys62pne

Track Everything: Limiting Prior Knowledge in Online Multi-Object Recognition

Sebastien C. Wong, Victor Stamatescu, Adam Gatt, David Kearney, Ivan Lee, Mark D. McDonnell
2017 IEEE Transactions on Image Processing  
An experimental evaluation demonstrates that our approach is competitive with state-of-the-art video object recognition systems that do make use of object-specific prior knowledge in detection and tracking  ...  is combined with object state information from the tracking algorithm.  ...  Object recognition then involves combining these class predictions, with state information given by object tracking.  ... 
doi:10.1109/tip.2017.2696744 pmid:28436874 fatcat:s7grj6ejxrcgjocc72mnaxjpxu

Learning Object Representations Using Sequential Patterns [chapter]

Nobuyuki Morioka
2008 Lecture Notes in Computer Science  
The temporal encoding represents the spatial relations between local features. We view the problem of object recognition as a sequential prediction task.  ...  Our method uses a Discriminative Variable Memory Markov (DVMM) model that precisely captures underlying characteristics of multiple statistical sources that generate sequential patterns in a stochastic  ...  National ICT Australia is funded by the Australian Government's Backing Australia's Ability initiative, in part through the Australian Research Council.  ... 
doi:10.1007/978-3-540-89378-3_56 fatcat:eaf5fkvq6rfptj3jpohs7ywp3q
« Previous Showing results 1 — 15 out of 759 results