STENT 发表于 2025-3-25 06:30:52

978-3-030-86336-4Springer Nature Switzerland AG 2021

Exposition 发表于 2025-3-25 09:51:55

http://reply.papertrans.cn/29/2824/282313/282313_22.png

white-matter 发表于 2025-3-25 15:02:53

http://reply.papertrans.cn/29/2824/282313/282313_23.png

PLUMP 发表于 2025-3-25 19:00:34

http://reply.papertrans.cn/29/2824/282313/282313_24.png

撕裂皮肉 发表于 2025-3-25 22:12:50

Fast Text vs. Non-text Classification of Imagestion is based on a block-level approach. FTC achieves 94.2% F-measure, 0.97 area under the ROC curve, and 74.8 ms and 8.6 ms inference times for CPU and GPU, respectively. A dataset of 1M images, automatically annotated with masks indicating text presence, is introduced and made public at ..

thrombosis 发表于 2025-3-26 02:26:33

FEDS - Filtered Edit Distance Surrogatevarious challenging scene text datasets such as IIIT-5K, SVT, ICDAR, SVTP, and CUTE. The proposed method provides an average improvement of . on total edit distance and an error reduction of . on accuracy.

modifier 发表于 2025-3-26 06:16:57

VML-HP: Hebrew Paleography Datasetve evaluated several deep learning classifiers on both of the test sets. The results show that convolutional networks can classify Hebrew script sub-types on a typical test set with accuracy much higher than the accuracy on the blind test.

Leisureliness 发表于 2025-3-26 10:30:43

http://reply.papertrans.cn/29/2824/282313/282313_28.png

Banister 发表于 2025-3-26 15:32:06

http://reply.papertrans.cn/29/2824/282313/282313_29.png

Hot-Flash 发表于 2025-3-26 20:46:54

http://reply.papertrans.cn/29/2824/282313/282313_30.png
页: 1 2 [3] 4 5 6
查看完整版本: Titlebook: Document Analysis and Recognition – ICDAR 2021; 16th International C Josep Lladós,Daniel Lopresti,Seiichi Uchida Conference proceedings 202