Filters








224 Hits in 6.3 sec

Document image decoding using Iterated Complete Path search with subsampled heuristic scoring

D.S. Bloomberg, T.P. Minka, K. Popat
Proceedings of Sixth International Conference on Document Analysis and Recognition  
It has been shown that the computation time of Document Image Decoding can be significantly reduced by employing heuristics in the search for the best decoding of a text line.  ...  We present the optimal (lowest upperbound) heuristic for any degree of subsampling of multilevel template and/or interpolation, for use in text line decoding with ICP.  ...  In this introduction, we give a quick review of the fundamental parameters and equations for a DID line decoder, and then describe the Iterated Complete Path (ICP) algorithm.  ... 
doi:10.1109/icdar.2001.953811 dblp:conf/icdar/BloombergPM01 fatcat:dsi5zvfizngetdivwz5tl54zbu

Document image decoding using iterated complete path search

Thomas P. Minka, Dan S. Bloomberg, Kris Popat, Paul B. Kantor, Daniel P. Lopresti, Jiangying Zhou
2000 Document Recognition and Retrieval VIII  
The computation time of Document Image Decoding can be significantly reduced by employing heuristics in the search for the best decoding of a text line.  ...  In the Iterated Complete Path method, template matches are performed only along the best path found by dynamic programming on each iteration.  ...  ACKNOWLEDGEMENTS This work is a direct outgrowth of the pioneering work by Gary Kopec and Phil Chou, and we wish to dedicate it to Gary, who is no longer with us to share in the fruits of his creativity  ... 
doi:10.1117/12.410843 dblp:conf/drr/MinkaBP01 fatcat:wvtupuranbefpnrasdw56sbqpm

Adding linguistic constraints to document image decoding: comparing the iterated complete path and stack algorithms

Kris Popat, Daniel H. Greene, Justin K. Romberg, Dan S. Bloomberg, Paul B. Kantor, Daniel P. Lopresti, Jiangying Zhou
2000 Document Recognition and Retrieval VIII  
Beginning with an observed document image and a model of how the image has been degraded, Document Image Decoding recognizes printed text by attempting to find a most probable path through a hypothesized  ...  The first, called the iterated complete path algorithm, involves iteratively rescoring complete paths using conditional language model probability distributions of increasing order, expanding state only  ...  ACKNOWLEDGMENTS This work has benefitted from discussions with Henry Baird, Thorsten Brants, Tom Breuel, Francine Chen, Gary Kopec, Tom Minka, and Les Niles.  ... 
doi:10.1117/12.410844 dblp:conf/drr/PopatGRB01 fatcat:hq77qflv2vdytnckz3wsf4mjw4

Document Recognition without Strong Models

Henry S. Baird
2011 2011 International Conference on Document Analysis and Recognition  
Can a high-performance document image recognition system be built without detailed knowledge of the application?  ...  When we can't collect (and label) enough real training data, does it help to complement them with data synthesized using generative models? Is it ever completely safe to rely on synthetic data?  ...  In response to this problem PARC researchers invented a procedure (iterated complete path optimization) to find an optimal path that is, on average, far faster [45] and compared it with an approximate  ... 
doi:10.1109/icdar.2011.91 dblp:conf/icdar/Baird11 fatcat:3mc4igzc3vd2llksd2hafvdtf4

Dynamic Boundary Time Warping for Sub-sequence Matching with Few Examples [article]

Łukasz Borchmann, Dawid Jurkiewicz, Filip Graliński, Tomasz Górecki
2020 arXiv   pre-print
Instead, we use query examples as is, utilizing all of them simultaneously.  ...  We are the first to propose an algorithm for such a search that does not rely on computing the average sequence from query examples.  ...  As shown by Wang and Jiang (1994) , multiple sequence alignment with the sum of all pairs score 2 is an NP-complete problem.  ... 
arXiv:2010.14464v1 fatcat:iq4liqjfxzbctkrvdocrswi4mm

High quality document image compression with "DjVu"

Le´on Bottou
1998 Journal of Electronic Imaging (JEI)  
Then, several novel techniques are used to maximize the compression ratio: the bi-level foreground image is encoded with AT&T's proposal to the new JBIG2 fax standard, and a new wavelet-based compression  ...  We present a new image compression technique called \DjVu " that is speci cally geared towards the compression of high-resolution, high-quality images of scanned documents in color.  ...  Document Image Compression with DjVu As we stated earlier, the digital library experience cannot be complete without a way of transmitting and displaying document images in color.  ... 
doi:10.1117/1.482609 fatcat:zxve2ne2jjgzpbdmlftbtzgmc4

Minkowski's Inequality Based Sensitivity Analysis of Fuzzy Signatures

István Á. Harmati, Ádám Bukovics, László T. Kóczy
2016 Journal of Artificial Intelligence and Soft Computing Research  
Finally, we apply our results to a fuzzy signature used in civil enginnering.  ...  In this paper we discuss the sensitivity of the weigthed general mean aggregation operator to the uncertainty of the input values, then we analyse the sensitivity of fuzzy signatures equipped with these  ...  A t-test (with the significance level p=0.01) was used to compare these results with the F-score values obtained with the ensemble of classifiers in Experiment 4 (without any subsample exchange tech-nique  ... 
doi:10.1515/jaiscr-2016-0016 fatcat:tpmfrlnp3bhbznaqxztvl4hray

Graph-Based Keyword Spotting in Historical Documents Using Context-Aware Hausdorff Edit Distance

Michael Stauffer, Andreas Fischer, Kaspar Riesen
2018 2018 13th IAPR International Workshop on Document Analysis Systems (DAS)  
The decoded string can further be used as a query for database search, e.g. in document retrieval.  ...  A language model can be used to improve decoding of the network output. Transcription, decoding and dictionary search are often seen as separate steps.  ... 
doi:10.1109/das.2018.31 dblp:conf/das/Stauffer0R18 fatcat:2r2cjpiitfcs5knjtqbfvcuwsi

On the applications of multimedia processing to communications

R.V. Cox, B.G. Haskell, Y. LeCun, B. Shahraray, L. Rabiner
1998 Proceedings of the IEEE  
It provides access to distributed data bases and, via some excellent search engines, has excellent search capabilities.  ...  With the emergence of the multimedia PC and its concomitant signal-processing capability, a simple solution was proposed, namely, the compression (and decompression) and coding (and decoding) of voice  ...  image processing.  ... 
doi:10.1109/5.664272 fatcat:yhn4kb3mmbctbbfus7br26oox4

Generative chemistry: drug discovery with deep learning generative models [article]

Yuemin Bian, Xiang-Qun Xie
2020 arXiv   pre-print
From the generation of original texts, images, and videos, to the scratching of novel molecular structures, the incredible creativity of deep learning generative models surprised us about the height machine  ...  This review starts with a brief history of artificial intelligence in drug discovery to outline this emerging paradigm.  ...  Notably, the complete de novo drug design cycle can be achieved with target prediction models for scoring.  ... 
arXiv:2008.09000v1 fatcat:ivznoc4bsbfoderwr2ted76fiq

NeurIPS 2020 NLC2CMD Competition: Translating Natural Language to Bash Commands [article]

Mayank Agarwal, Tathagata Chakraborti, Quchen Fu, David Gros, Xi Victoria Lin, Jaron Maene, Kartik Talamadupula, Zhongwei Teng, Jules White
2021 arXiv   pre-print
This is a report on the competition with details of the task, metrics, data, attempted solutions, and lessons learned.  ...  Participants were tasked with building models that can transform descriptions of command line tasks in English to their Bash syntax.  ...  Scores in the beam search were produced using an approximation 21 for confidence due to the high correlation observed between correct predictions and high beam scores. 3.  ... 
arXiv:2103.02523v2 fatcat:2wva6wca5jbhrfslqej5efj5wi

Advances in Electron Microscopy with Deep Learning

Jeffrey Ede
2020 Zenodo  
This doctoral thesis covers some of my advances in electron microscopy with deep learning.  ...  , uniformly spaced and other fixed sparse scan paths; recurrent neural networks trained to piecewise adapt sparse scan paths to specimens by reinforcement learning; improving signal-to-noise; and conditional  ...  In addition, part of the text in section 1.2 is adapted from our earlier work with permission 201 under a Creative Commons Attribution 4.0 73 license.  ... 
doi:10.5281/zenodo.4598227 fatcat:hm2ksetmsvf37adjjefmmbakvq

Advances in Electron Microscopy with Deep Learning

Jeffrey Ede
2020 Zenodo  
This doctoral thesis covers some of my advances in electron microscopy with deep learning.  ...  , uniformly spaced and other fixed sparse scan paths; recurrent neural networks trained to piecewise adapt sparse scan paths to specimens by reinforcement learning; improving signal-to-noise; and conditional  ...  In addition, part of the text in section 1.2 is adapted from our earlier work with permission 201 under a Creative Commons Attribution 4.0 73 license.  ... 
doi:10.5281/zenodo.4591029 fatcat:zn2hvfyupvdwlnvsscdgswayci

Advances in Electron Microscopy with Deep Learning

Jeffrey Ede
2020 Zenodo  
This doctoral thesis covers some of my advances in electron microscopy with deep learning.  ...  , uniformly spaced and other fixed sparse scan paths; recurrent neural networks trained to piecewise adapt sparse scan paths to specimens by reinforcement learning; improving signal-to-noise; and conditional  ...  In addition, part of the text in section 1.2 is adapted from our earlier work with permission 201 under a Creative Commons Attribution 4.0 73 license.  ... 
doi:10.5281/zenodo.4399748 fatcat:63ggmnviczg6vlnqugbnrexsgy

Advances in Electron Microscopy with Deep Learning

Jeffrey Ede
2020 Zenodo  
This doctoral thesis covers some of my advances in electron microscopy with deep learning.  ...  , uniformly spaced and other fixed sparse scan paths; recurrent neural networks trained to piecewise adapt sparse scan paths to specimens by reinforcement learning; improving signal-to-noise; and conditional  ...  In addition, part of the text in section 1.2 is adapted from our earlier work with permission 201 under a Creative Commons Attribution 4.0 73 license.  ... 
doi:10.5281/zenodo.4413249 fatcat:35qbhenysfhvza2roihx52afuy
« Previous Showing results 1 — 15 out of 224 results