A real-time multitarget tracking system with robust multichannel CNN-UM algorithms

Timar, G.; Rekeczky, Csaba

doi:10.1109/tcsi.2005.851703

Cited by 17 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, multiple object tracking and data evaluation tasks are carried out in the post-processing part. This architecture is similar to that of proposed in [6].…”

Section: Tracking Algorithm and System Architecturesupporting

confidence: 59%

See 1 more Smart Citation

Realization of preprocessing blocks of CNN based CASA system on FPGA

Savkay

Yildiz

Cesur

et al. 2013

2013 European Conference on Circuit Theory and Design (ECCTD)

View full text Add to dashboard Cite

In this paper, hardware optimization of the preprocessing part of a computer aided semen analysis (CASA) system is proposed, which is also implemented on an FPGA device as a working prototype. A real-time cellular neural network (CNN) emulator (RTCNNP-v2) is used for the realization of the image processing algorithms, whose regular, flexible and reconfigurable infrastructure simplifies the prototyping process. For future work, the post-processing part of the CASA system is proposed to be implemented on the same FPGA device as software, using either a soft or hard processor core. By the integration of the pre-and post-processing parts, the designed CASA system will be capable of processing full-HD 1080p@60 (1080×1920) video images in real-time.

show abstract

“…On the other hand, multiple object tracking and data evaluation tasks are carried out in the post-processing part. This architecture is similar to that of proposed in [6].…”

Section: Tracking Algorithm and System Architecturesupporting

confidence: 59%

“…Viewed from an engineering viewpoint, this system is principally a multiple moving object tracking system (MMOTS) [5], [6], specifically designed for sperm motility analysis. The preprocessing blocks are implemented on a previously proposed CNN emulator [7], [8], [9], which was also realized on various FPGA devices.…”

Section: Introductionmentioning

confidence: 99%

Realization of preprocessing blocks of CNN based CASA system on FPGA

Savkay

Yildiz

Cesur

et al. 2013

2013 European Conference on Circuit Theory and Design (ECCTD)

View full text Add to dashboard Cite

show abstract

“…There, also, are various related tracking research using various neural networks. For example, neural networks have been used for human tracking [11], recognition of hand gestures [12], pedestrian detection [13], robust tracking with a single modified CNN [14], multitarget tracking [15], [16], and a tracking method combining CNNs and optical flow [17]. Deconvolutional networks and a FCNs have been used for segmentation [18], [19].…”

Section: Related Workmentioning

confidence: 99%

Globally Optimal Object Tracking with Fully Convolutional Networks

Lee¹,

Iwana²,

Ide³

et al. 2016

Preprint

View full text Add to dashboard Cite

Tracking is one of the most important but still difficult tasks in computer vision and pattern recognition. The main difficulties in the tracking field are appearance variation and occlusion. Most traditional tracking methods set the parameters or templates to track target objects in advance and should be modified accordingly. Thus, we propose a new and robust tracking method using a Fully Convolutional Network (FCN) to obtain an object probability map and Dynamic Programming (DP) to seek the globally optimal path through all frames of video. Our proposed method solves the object appearance variation problem with the use of a FCN and deals with occlusion by DP. We show that our method is effective in tracking various single objects through video frames.

show abstract

“…The weights are arranged into ‘templates’ of a form analogous to convolution operators that are widely used in digital‐image processing. As a result, image and video processing is the main considered domain of CNN applications, and several methods, ranging from simple image filtering tasks through complex, nonlinear image analysis algorithms , have been proposed for CNNs so far.…”

Section: Introductionmentioning

confidence: 99%

Fast texture synthesis with cellular neural network‐based patch stitching

Ślot

Kornatowski²,

Dębiec

2012

Circuit Theory & Apps

View full text Add to dashboard Cite

The presented paper introduces a novel approach to patch‐based texture synthesis. The proposed method utilizes Cellular Neural Networks (CNNs) for patch stitching, the most important operation of the synthesis procedure, expected to transform a patchwork made up of pieces of some small texture sample into a seamless, large texture. To fill in between‐patch gaps, spontaneous pattern formation capabilities of CNNs are used. By using autoregressive filtering framework, we show that CNNs can be designed to generate complex stochastic fields that approximate stochastic and near‐stochastic textures. The proposed texture‐synthesis procedure is the following. After creating a patchwork from randomly drawn pieces of a small target texture, we fix boundary CNN conditions to edge pixels of all patches subject to stitching and execute an appropriately derived texture‐approximating template. Finally, brightness of the resulting patterns is locally adjusted by executing a diffusion CNN template. As both processing steps involve pure CNN dynamics, the proposed patch‐stitching method can be extremely fast if implemented in existing hardware (at the order of milliseconds for synthesizing several megapixel images). Copyright © 2012 John Wiley & Sons, Ltd.

show abstract

A real-time multitarget tracking system with robust multichannel CNN-UM algorithms

Cited by 17 publications

References 21 publications

Realization of preprocessing blocks of CNN based CASA system on FPGA

Realization of preprocessing blocks of CNN based CASA system on FPGA

Globally Optimal Object Tracking with Fully Convolutional Networks

Fast texture synthesis with cellular neural network‐based patch stitching

Contact Info

Product

Resources

About