11,005 Hits in 7.6 sec

Real-Time Object Segmentation in Image Sequences
연속 영상 기반 실시간 객체 분할

Eui-Seon Kang, Seung-Hun Yoo
2011 The KIPS Transactions PartB  
This paper shows an approach for real-time object segmentation on GPU (Graphics Processing Unit) using CUDA (Compute Unified Device Architecture).  ...  Recently, many applications that is monitoring system, motion analysis, object tracking or etc require real-time processing. It is not suitable for object segmentation to procedure real-time in CPU.  ...  Finally, we can achieve real-time object segmentation on GPU using CUDA and provide the basis for real-time tracking or behavior analysis.  ... 
doi:10.3745/kipstb.2011.18b.4.173 fatcat:k5wzzc7z2vgaxjzz5t7pmfrpfq

Tensor-Based CUDA Optimization for ANN Inferencing Using Parallel Acceleration on Embedded GPU [chapter]

Ahmed Khamis Abdullah Al Ghadani, Waleeja Mateen, Rameshkumar G. Ramaswamy
2020 IFIP Advances in Information and Communication Technology  
Be that as it may, Autonomous Mobile Robots (AMR) are Embedded Systems (ESs) with limited on-board resources. Thus, efficient techniques in ANN inferencing are required for real-time performance.  ...  These findings will contribute to ES developers and industries will benefit from real-time performance inferencing for AMR automation solutions.  ...  YOLO ANN for Object Detection YOLO ANN implements a smart approach for object detection. Achieving real-time performance using a single Convolutional Neural Network (CNN) structure.  ... 
doi:10.1007/978-3-030-49161-1_25 fatcat:kijobfe7j5bv7cca57rcluqgd4

An Intelligent Image Processing System for Real-Time Detection of Surface Flaws

Chou Yu-Cheng
2016 International Journal of Automation and Smart Technology  
A parallel prototype system, consisting of three CPUs and two GPUs, is used to inspect a simulated object with an area eight times greater than that of the real object in our previous work.  ...  This paper presents an intelligent parallel image processing system, PVCS (Parallel Visual Computing System), to handle in-line surface defect detection for large objects.  ...  A single CUDA workstation ideally needs to spend requires 50 time windows to process the entire object image for an , if the object is carried by on a platform capable of moving in the X and Y directions  ... 
doi:10.5875/ausmt.v6i1.1006 fatcat:6lj3gwgpbzbo3oq5setcjt3siq

Image and Video Processing on GPU: Implementation Scheme, Applications and Future Directions [chapter]

Roberto Di Salvo, Carmelo Pino
2013 Lecture Notes in Electrical Engineering  
In this paper the most recent image and video processing techniques and relevant applications are reviewed, highlighting the advantages of using CUDA in terms of efficiency.  ...  To this end, more recently the general-purpose computing on graphic processing units (GPGPUs) architectures, which are becoming increasingly programmable and scalable, have been adopted.  ...  Lillywhite et al. in [49] propose a similar real-time human detection approach using histograms of oriented gradients method through the NVIDIA CUDA architecture on a GPU.  ... 
doi:10.1007/978-3-642-31528-2_59 fatcat:gh7hxku6s5cffjojwud3x6fkli

A TBB-CUDA Implementation for Background Removal in a Video-Based Fire Detection System

Fan Wang, Xiao Jiang, Xiao Peng Hu
2014 Mathematical Problems in Engineering  
This implementation can exploit the combined computation power of TBB-CUDA, which can be applied to the real-time environment. Over 220 video sequences are utilized in the experiments.  ...  In this framework, TBB mainly deals with initializing work of the estimated Gaussian model running on CPU, and CUDA performs background removal and adaption of the model running on GPU.  ...  data transfer time on GPU.  ... 
doi:10.1155/2014/692921 fatcat:uw56wz4mzbbqhbb3ntrmmwsx64

Movement Tracking in Terrain Conditions Accelerated with CUDA

Piotr Skłodowski, Witold Żorski
2014 Proceedings of the 2014 Federated Conference on Computer Science and Information Systems  
In order to obtain a real-time system the CUDA technology was involved.  ...  The presented solution is based on well known image processing methods, nevertheless the key was the use of an appropriate conduct procedure.  ...  In order to obtain a real-time system the CUDA technology was involved.  ... 
doi:10.15439/2014f282 dblp:conf/fedcsis/SklodowskiZ14 fatcat:c2nwdalverfbtd6c5o7jcs2vde

Moving object detection strategy for augmented-reality applications in a GPGPU by using CUDA

Daniel Berjon, Carlos Cuevas, Francisco Moran, Narciso Garcia
2012 2012 IEEE International Conference on Consumer Electronics (ICCE)  
This strategy is suitable for augmented-reality applications, providing real-time high-quality results in a great variety of scenarios.  ...  A spatial-color-based non-parametric backgroundforeground modeling strategy in a GPGPU by using CUDA is proposed.  ...  In this paper we propose a novel non-parametric based moving object detection strategy in a GPGPU by using CUDA (Compute Unified Device Architecture) [7] , which provides real-time high-quality results  ... 
doi:10.1109/icce.2012.6161886 dblp:conf/iccel/BerjonCMG12 fatcat:yhpd5fnik5bmxc2ppmar4tv4ea

Adapting to a Cambrian AI/SW/HW explosion with open co-design competitions and Collective Knowledge

Grigori Fursin
2017 Zenodo  
detection object detection ImageNet ImageNet Car video stream Car video stream Real surveillance camera Real surveillance camera GEMM OpenCL GEMM OpenCL convolution CPU convolution CPU  ...  detection object detection ImageNet ImageNet Car video stream Car video stream Real surveillance camera Real surveillance camera GEMM OpenCL GEMM OpenCL convolution CPU convolution CPU  ... 
doi:10.5281/zenodo.2544257 fatcat:76ikgqtuhfhpzpywyo5h5xnclu

Application of graphics processing units to search pipelines for gravitational waves from coalescing binaries of compact objects

Shin Kee Chung, Linqing Wen, David Blair, Kipp Cannon, Amitava Datta
2010 Classical and quantum gravity  
We show that substantial improvements are possible and discuss the reduction in CPU count required for the detection of inspiral sources afforded by the use of GPUs.  ...  We report a novel application of graphics processing units (GPUs) for the purpose of accelerating the search pipelines for gravitational waves from coalescing binaries of compact objects.  ...  Much faster processing is therefore required for real-time detection and determination of source directions.  ... 
doi:10.1088/0264-9381/27/13/135009 fatcat:mcx4eaictzbahjsnihkc3malbq

Multi-GPU based framework for real-time motion analysis and tracking in multi-user scenarios

Sidi Ahmed Mahmoudi
2015 EAI Endorsed Transactions on Creative Technologies  
Based on this framework, we developed several real-time and GPU based video processing applications such as motion detection using moving camera, event detection and event localization  ...  As result, the current implementations, even running on modern hardware, cannot provide a real-time processing (25 frames per second, fps).  ...  Thus, the current implementations, even running on modern hardware, cannot provide a real-time processing.  ... 
doi:10.4108/ct.2.2.e5 fatcat:izmwhk6bwrekbi33zwhdwyjrfe

CUDA-based real-time face recognition system

Ren Meng, Zhang Shengbing, Lei Yi, Zhang Meng
2014 2014 Fourth International Conference on Digital Information and Communication Technology and its Applications (DICTAP)  
This paper proposes a real-time face recognition system based on the Compute Unified Device Architecture (CUDA) platform, which effectively completed the face detection and recognition tasks.  ...  It shows that the system we proposed achieves a good real-time performance.  ...  The fast real-time object detection method -Viola-Jones cascade classifier [1] was proposed by Viola and Jones from the University of Cambridge in 2001.  ... 
doi:10.1109/dictap.2014.6821688 dblp:conf/dictap/MengZYM14 fatcat:fcqfchjeyvbbzdnliotm4ipvri

Algorithm of search and track of static and moving large-scale objects

Anatoly Kalyaev, Maxim Khisamutdinov, L. Long, Y. Li, X. Li, Y. Dai, H. Yang
2017 ITM Web of Conferences  
The possible software implementation of the algorithm, based on multithread CUDA processing, is suggested. Experimental analysis of the suggested algorithm implementation is performed.  ...  We suggest an algorithm for processing of a sequence, which contains images of search and track of static and moving large-scale objects.  ...  The execution time is measured in seconds. Figure 1 . 1 Detecting and tracking of static large-scale objects. Figure 2 . 2 Detecting and tracking of moving large-scale objects.  ... 
doi:10.1051/itmconf/20171205014 fatcat:fycpwmrtqze2rjbuhtmi3rg5s4

Accelerating open science and AI with automated, portable, customizable and reusable research components and workflows

Grigori Fursin
2018 Zenodo  
In this presentation, we will introduce CK concepts and present several real-world use cases from General Motors and Arm on collaborative benchmarking, autotuning and co-design of efficient software/hardware  ...  interactive articles, and so on:  ...  detection object detection ImageNet ImageNet Car video stream Car video stream Real surveillance camera Real surveillance camera GEMM OpenCL GEMM OpenCL convolution CPU convolution CPU  ... 
doi:10.5281/zenodo.2544261 fatcat:433asnbyhjdgziacjagakcz2ki

Passive Radar Parallel Processing Using General-Purpose Computing on Graphics Processing Units

Karolina Szczepankiewicz, Mateusz Malanowski, Michał Szczepankiewicz
2015 International Journal of Electronics and Telecommunications  
As the computational load associated with passive radar processing is very high, NVIDIA CUDA technology has been employed for effective implementation using parallel processing.  ...  It is focused on constructing passive radar that works in real time.  ...  The obtained results clearly show that Matlab implementation is not suitable for real-time operations.  ... 
doi:10.1515/eletel-2015-0047 fatcat:vnpb7iuv6rggdo3bwpqqqraz24

Accelerating the local outlier factor algorithm on a GPU for intrusion detection systems

Malak Alshawabkeh, Byunghyun Jang, David Kaeli
2010 Proceedings of the 3rd Workshop on General-Purpose Computation on Graphics Processing Units - GPGPU '10  
We also consider the impact of input data set size, the neighborhood size (i.e., the value of k) and the feature space dimension, and report on their impact on execution time.  ...  The algorithm defines the notion of local outlier in which the degree to which an object is outlying is dependent on the density of its local neighborhood, and each object can be assigned an LOF which  ...  Of course, near real-time performance is needed in order to detect any malicious activity before the system becomes fully compromised.  ... 
doi:10.1145/1735688.1735707 dblp:conf/asplos/AlshawabkehJK10 fatcat:afkaslln4nekpohayxoxjsat4e
« Previous Showing results 1 — 15 out of 11,005 results