Filters








929 Hits in 8.1 sec

Self-Supervised Material and Texture Representation Learning for Remote Sensing Tasks [article]

Peri Akiva, Matthew Purri, Matthew Leotta
2021 arXiv   pre-print
It is often used as a precursor step to obtain useful initial network weights which contribute to faster convergence and superior performance of downstream tasks.  ...  land cover classification, and semantic segmentation tasks.  ...  vector z i and learned cluster center q υ using r 1×D i,υ = z i − q υ .  ... 
arXiv:2112.01715v1 fatcat:ajgy3cwm6fbbhn7y7uepetsfmu

Phase synchrony facilitates binding and segmentation of natural images in a coupled neural oscillator network

Holger Finger, Peter König
2014 Frontiers in Computational Neuroscience  
This demonstrates that the concept of binding by synchrony is applicable in the context of unsupervised learning using natural visual stimuli.  ...  Here we investigate this concept based on unsupervised learning using natural visual stimuli. We simulate dual-variable neural oscillators with separate activation and phase variables.  ...  ACKNOWLEDGMENTS The authors would like to thank Robert Märtin for his valuable comments and helpful suggestions.  ... 
doi:10.3389/fncom.2013.00195 pmid:24478685 pmcid:PMC3902207 fatcat:gqovw4eqvreklp4ybhoxmghory

Saliency-driven unstructured acoustic scene classification using latent perceptual indexing

Ozlem Kalinli, Shiva Sundaram, Shrikanth Narayanan
2009 2009 IEEE International Workshop on Multimedia Signal Processing  
Similar to latent semantic indexing of text documents, the classification system uses unit-document frequency measure to index the clip in a continuous, latent space.  ...  Motivated by the bottom-up attention model of the human auditory system, salient events of an audio clip are extracted in an unsupervised manner and presented to the classification system.  ...  Center-surround differences are computed as point wise differences across scales using three center scales c = {2, 3, 4} and two surround scales s = c + δ with δ {3, 4} resulting in six feature maps for  ... 
doi:10.1109/mmsp.2009.5293267 dblp:conf/mmsp/KalinliSN09 fatcat:5y2m36c57bgs5azzj4lnvg3yre

Probabilistic Semantic Segmentation Refinement by Monte Carlo Region Growing [article]

Philipe A. Dias, Henry Medeiros
2020 arXiv   pre-print
Exploiting concepts of Bayesian estimation and variance reduction techniques, pRGR performs multiple refinement iterations at varied receptive fields sizes, while updating cluster statistics to adapt to  ...  Semantic segmentation with fine-grained pixel-level accuracy is a fundamental component of a variety of computer vision applications.  ...  receptive fields.  ... 
arXiv:2005.05856v1 fatcat:xh7dn34ekrhn3md6h2qdijw65y

Image Perception

Gloria Menegaz, Guang-Zhong Yang
2007 EURASIP Journal on Advances in Signal Processing  
denoising step with surround inhibition at each level of multiscale image decomposition to solve the problem of oversegmentation which affects classical edge detectors in the presence of textures.  ...  processes while a stimulus is present," and to establish associated computational models that can be generalized and exploited for designing a human-centered approach to imaging.  ...  An unsupervised segmentation step based on fuzzy C-means clustering is employed to partition the input image into a suitable number of segments.  ... 
doi:10.1155/2007/39068 fatcat:vjo65k6zmrgoledczrixyuc4ru

Deep Learning-based Action Detection in Untrimmed Videos: A Survey [article]

Elahe Vahdani, Yingli Tian
2021 arXiv   pre-print
This paper provides an extensive overview of deep learning-based algorithms to tackle temporal action detection in untrimmed videos with different supervision levels including fully-supervised, weakly-supervised  ...  Despite the progress of action recognition algorithms in trimmed videos, the majority of real-world videos are lengthy and untrimmed with sparse segments of interest.  ...  Then, the receptive field of each anchor segment was aligned with its temporal span using dilated temporal convolutions. This idea was also used in TSA-Net [33] .  ... 
arXiv:2110.00111v1 fatcat:ven4rijqmnbyxflrf6wyxfpex4

Joint Multi-Image Saliency Analysis for Region of Interest Detection in Optical Multispectral Remote Sensing Images

Jie Chen, Libao Zhang
2016 Remote Sensing  
First, bisecting K-means clustering on the entire image set allows us to extract the global correspondence among multiple images in RGB and CIELab color spaces.  ...  multiple images with respect to certain recurring patterns, e.g., intensity, color, texture, or shape.  ...  global contrast to simulate human visual receptive fields.  ... 
doi:10.3390/rs8060461 fatcat:xojbomduxfggfhoh52et2plqta

Self-Organizing Map-Based Color Image Segmentation with k-Means Clustering and Saliency Map

Dongxiang Chi
2011 ISRN Signal Processing  
In SOM-K, pixel features of intensity and L∗u∗v∗ color space are trained with SOM and followed by a k-means method to cluster the prototype vectors, which are filtered with hits map.  ...  It is observed that SOM-K and SOM-KS, being an unsupervised method, can achieve better segmentation results with less computational load and no human intervention.  ...  of their receptive fields, neurons are excited by one color and inhibited by another color, while the converse is true in the surround.  ... 
doi:10.5402/2011/393891 fatcat:opdgc4tznzbuhpdpjallbva6qq

Salient Object Detection: A Survey [article]

Ali Borji, Ming-Ming Cheng, Qibin Hou, Huaizu Jiang, Jia Li
2018 arXiv   pre-print
We aim to provide a comprehensive review of the recent progress in salient object detection and situate this field among other closely related areas such as generic scene segmentation, object proposal  ...  Detecting and segmenting salient objects in natural scenes, often referred to as salient object detection, has attracted a lot of interest in computer vision.  ...  With the colorized motion field as the input image, the local multi-scale contrast, regional center-surround distance, and global spatial distribution are computed and finally integrated in a linear way  ... 
arXiv:1411.5878v5 fatcat:aw7pqd554zdfblokxy2zgp35z4

Salient object detection: A survey

Ali Borji, Ming-Ming Cheng, Qibin Hou, Huaizu Jiang, Jia Li
2019 Computational Visual Media  
We aim to provide a comprehensive review of recent progress in salient object detection and situate this field among other closely related areas such as generic scene segmentation, object proposal generation  ...  Detecting and segmenting salient objects from natural scenes, often referred to as salient object detection, has attracted great interest in computer vision.  ...  Neurons with large receptive fields provide global information that can help better identify the most salient region in an image, while neurons with small receptive fields provide local information that  ... 
doi:10.1007/s41095-019-0149-9 fatcat:rwh3mzauinfj7na5ewo6wfp26e

Fast Regions-of-Interest Detection in Whole Slide Histopathology Images [chapter]

Junzhou Huang, Ruoyu Li
2020 Histopathology and Liquid Biopsy [Working Title]  
To efficiently construct superpixels with fine details preserved, we utilized a novel superpixel clustering algorithm which cluster blocks of pixel in a hierarchical fashion.  ...  The former maintains the accuracy of segmentation, meanwhile, avoids most of unnecessary revisit to the 'non-boundary' pixels.  ...  Therefore, traditional fully convolutional networks, used to work perfectly for medical image segmentation [8] , are no longer applicable, because of the parameter scale that may explode and the rising  ... 
doi:10.5772/intechopen.94238 fatcat:mknyksvgq5fffd6cmvsivxnmzi

Deep Learning in Image Cytometry: A Review

Anindya Gupta, Philip J. Harrison, Håkan Wieslander, Nicolas Pielawski, Kimmo Kartasalo, Gabriele Partel, Leslie Solorzano, Amit Suveer, Anna H. Klemm, Ola Spjuth, Ida‐Maria Sintorn, Carolina Wählby
2018 Cytometry Part A  
Starting with an analogy to neuroscience, we aim to give the reader an overview of the key concepts of neural networks, and an understanding of how deep learning differs from more classical approaches  ...  In this review, we focus on deep learning and how it is applied to microscopy image data of cells and tissue samples.  ...  We are earnestly grateful to Emeritus Prof. Ewert Bengtsson, and Petter Ranefall for their appreciative suggestions. LITERATURE CITED  ... 
doi:10.1002/cyto.a.23701 pmid:30565841 pmcid:PMC6590257 fatcat:dszbcsfncrhxnazsxopjkbe3ju

Bayesian Saliency via Low and Mid Level Cues

Yulin Xie, Huchuan Lu, Ming-Hsuan Yang
2013 IEEE Transactions on Image Processing  
We present a Laplacian sparse subspace clustering method to group superpixels with local features, and analyze the results with respect to the coarse saliency region to compute the prior saliency map.  ...  Visual saliency detection is a challenging problem in computer vision with great importance which finds numerous applications.  ...  In [25] , Achanta et al. exploit the center-surround principle by comparing color features of each pixel with average values of the whole image to compute the salience.  ... 
doi:10.1109/tip.2012.2216276 pmid:22955904 fatcat:5upmuvkcyzdbjgwgzaiyhxqiq4

Learning Rich Representations For Structured Visual Prediction Tasks [article]

Mohammadreza Mostajabi
2019 arXiv   pre-print
When used in conjunction with modern neural architectures such as ResNet, DenseNet and NASNet (to which it is complementary) our approach achieves competitive accuracy on segmentation benchmarks.  ...  Applied to semantic segmentation and other structured prediction tasks, our approach exploits statistical structure in the image and in the label space without setting up explicit structured prediction  ...  (including background) classifier network, with receptive field of 1×1.  ... 
arXiv:1908.11820v1 fatcat:n2utrggy5faszodf4noe5ayram

A neural computational model for bottom-up attention with invariant and overcomplete representation

Zou Qi, Zhao Songnian, Wang Zhe, Huang Yaping
2012 BMC Neuroscience  
For example, Treisman and Gelade [1] developed the feature integration theory to explain how primary visual features are processed and represented with separate feature maps and are later integrated into  ...  Numerous studies have explored theories and computational models to provide an efficient input to the saliency detection.  ...  We would like to thank S. Sarkar for invaluable help and suggestions. We greatly appreciate the reviewers for their suggestions. Author details  ... 
doi:10.1186/1471-2202-13-145 pmid:23190754 pmcid:PMC3599588 fatcat:bjwe4evwavhl3n3swavspw5j2m
« Previous Showing results 1 — 15 out of 929 results