Matches in SemOpenAlex for { <https://semopenalex.org/work/W2890140353> ?p ?o ?g. }
- W2890140353 abstract "This work presents a method for visual text recognition without using any paired supervisory data. We formulate the text recognition task as one of aligning the conditional distribution of strings predicted from given text images, with lexically valid strings sampled from target corpora. This enables fully automated, and unsupervised learning from just line-level text-images, and unpaired text-string samples, obviating the need for large aligned datasets. We present detailed analysis for various aspects of the proposed method, namely - (1) impact of the length of training sequences on convergence, (2) relation between character frequencies and the order in which they are learnt, (3) generalisation ability of our recognition network to inputs of arbitrary lengths, and (4) impact of varying the text corpus on recognition accuracy. Finally, we demonstrate excellent text recognition accuracy on both synthetically generated text images, and scanned images of real printed books, using no labelled training examples." @default.
- W2890140353 created "2018-09-27" @default.
- W2890140353 creator A5057678172 @default.
- W2890140353 creator A5060511349 @default.
- W2890140353 creator A5086192896 @default.
- W2890140353 date "2018-09-23" @default.
- W2890140353 modified "2023-09-27" @default.
- W2890140353 title "Learning to Read by Spelling: Towards Unsupervised Text Recognition." @default.
- W2890140353 cites W1491389626 @default.
- W2890140353 cites W1505401468 @default.
- W2890140353 cites W1521064364 @default.
- W2890140353 cites W1527085884 @default.
- W2890140353 cites W1533861849 @default.
- W2890140353 cites W1557952530 @default.
- W2890140353 cites W1569614731 @default.
- W2890140353 cites W1607307044 @default.
- W2890140353 cites W1647671624 @default.
- W2890140353 cites W1895191496 @default.
- W2890140353 cites W1903029394 @default.
- W2890140353 cites W1922126009 @default.
- W2890140353 cites W1975453746 @default.
- W2890140353 cites W1978729128 @default.
- W2890140353 cites W1980829015 @default.
- W2890140353 cites W1981283549 @default.
- W2890140353 cites W1998042868 @default.
- W2890140353 cites W2001642682 @default.
- W2890140353 cites W2049633694 @default.
- W2890140353 cites W2049951199 @default.
- W2890140353 cites W2053007711 @default.
- W2890140353 cites W2053317383 @default.
- W2890140353 cites W2061802763 @default.
- W2890140353 cites W2061822043 @default.
- W2890140353 cites W2069472161 @default.
- W2890140353 cites W2099471712 @default.
- W2890140353 cites W2112890726 @default.
- W2890140353 cites W2114781615 @default.
- W2890140353 cites W2115367993 @default.
- W2890140353 cites W2118696714 @default.
- W2890140353 cites W2122221966 @default.
- W2890140353 cites W2123595463 @default.
- W2890140353 cites W2126086996 @default.
- W2890140353 cites W2127141656 @default.
- W2890140353 cites W2133485975 @default.
- W2890140353 cites W2136346830 @default.
- W2890140353 cites W2147800946 @default.
- W2890140353 cites W2153182373 @default.
- W2890140353 cites W2163605009 @default.
- W2890140353 cites W2168868236 @default.
- W2890140353 cites W2179581459 @default.
- W2890140353 cites W2194775991 @default.
- W2890140353 cites W2252189001 @default.
- W2890140353 cites W2265364998 @default.
- W2890140353 cites W2335728318 @default.
- W2890140353 cites W2339754110 @default.
- W2890140353 cites W2481930807 @default.
- W2890140353 cites W2592480533 @default.
- W2890140353 cites W2605287558 @default.
- W2890140353 cites W2765961751 @default.
- W2890140353 cites W2766182427 @default.
- W2890140353 cites W2784107021 @default.
- W2890140353 cites W2949117887 @default.
- W2890140353 cites W2949888546 @default.
- W2890140353 cites W2950477116 @default.
- W2890140353 cites W2950561535 @default.
- W2890140353 cites W2950635152 @default.
- W2890140353 cites W2952302849 @default.
- W2890140353 cites W2952439627 @default.
- W2890140353 cites W2952453038 @default.
- W2890140353 cites W2953384591 @default.
- W2890140353 cites W2962790387 @default.
- W2890140353 cites W2963073614 @default.
- W2890140353 cites W2963517393 @default.
- W2890140353 cites W2964300754 @default.
- W2890140353 cites W2964308564 @default.
- W2890140353 cites W70975097 @default.
- W2890140353 hasPublicationYear "2018" @default.
- W2890140353 type Work @default.
- W2890140353 sameAs 2890140353 @default.
- W2890140353 citedByCount "2" @default.
- W2890140353 countsByYear W28901403532019 @default.
- W2890140353 crossrefType "posted-content" @default.
- W2890140353 hasAuthorship W2890140353A5057678172 @default.
- W2890140353 hasAuthorship W2890140353A5060511349 @default.
- W2890140353 hasAuthorship W2890140353A5086192896 @default.
- W2890140353 hasConcept C115961682 @default.
- W2890140353 hasConcept C138885662 @default.
- W2890140353 hasConcept C153180895 @default.
- W2890140353 hasConcept C154945302 @default.
- W2890140353 hasConcept C157486923 @default.
- W2890140353 hasConcept C162324750 @default.
- W2890140353 hasConcept C187736073 @default.
- W2890140353 hasConcept C204321447 @default.
- W2890140353 hasConcept C2524010 @default.
- W2890140353 hasConcept C2777801307 @default.
- W2890140353 hasConcept C2780451532 @default.
- W2890140353 hasConcept C2780861071 @default.
- W2890140353 hasConcept C28490314 @default.
- W2890140353 hasConcept C2983812711 @default.
- W2890140353 hasConcept C33923547 @default.
- W2890140353 hasConcept C37914503 @default.