Transformer-based Recognition
Transformer-Based Recognition is a paradigm in optical character recognition that employs transformer architectures to model long-range dependencies and contextual relationships in text imagery through self-attention mechanisms, distinguishing itself from convolutional or recurrent approaches by its global receptive field and parallel sequence processing.
330
Publications
81K
Citations
1.4K
Authors
386
Institutions
Publications per year
2017–2026
327
330
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov et al. · arXiv (Cornell University) · 2020 · 21.6K citations · Full text
Conformer: Convolution-augmented Transformer for Speech Recognition
Anmol Gulati, James Qin, Chung‐Cheng Chiu et al. · 2020 · 2.7K citations
Ze Liu, Ning Jia, Yue Cao et al. · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · 2022 · 1.9K citations
1–5 of 330
1.4K
5
h-index
4
h-index
4
h-index
4
h-index
4
h-index
1–5 of 1.4K
386
University of Chinese Academy of Sciences
Beijing, China
12 h-index
39 publications
2.3K citations
Google (United States)
Mountain View, United States
11 h-index
68 publications
255.6K citations
Microsoft Research Asia (China)
Beijing, China
10 h-index
34 publications
256K citations

Chinese Academy of Sciences
Beijing, China
9 h-index
37 publications
4.3K citations
Peng Cheng Laboratory
Shenzhen, China
9 h-index
14 publications
1.9K citations
Tsinghua University
Beijing, China
7 h-index
12 publications
36.9K citations
Beijing Academy of Artificial Intelligence
Beijing, China
7 h-index
17 publications
2.6K citations
University of Science and Technology of China
Hefei, China
7 h-index
18 publications
32.9K citations
1–8 of 386
Repository · 73 publications · 26K citations
33
h-index
Conference · 17 publications · 5K citations
17
h-index
8 publications · 280 citations
8
h-index
7 publications · 214 citations
7
h-index
Conference · 7 publications · 1.1K citations
7
h-index
1–5