1 |
Exploring the Impact of Negative Samples of Contrastive Learning: A Case Study of Sentence Embedding
|
|
|
|
BASE
|
|
Show details
|
|
2 |
Activated Epithelial FGF8 Signaling Induces Fused Supernumerary Incisors ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
Activated Epithelial FGF8 Signaling Induces Fused Supernumerary Incisors ...
|
|
|
|
BASE
|
|
Show details
|
|
4 |
A translated volume and its many covers: A Diachronic, Social-Semiotic Approach to the Study of Translated Book Covers
|
|
Li, L. - : Bloomsbury, 2021. : London and New York, 2021
|
|
BASE
|
|
Show details
|
|
7 |
Study of central exclusive [Image: see text] production in proton-proton collisions at [Formula: see text] and 13TeV
|
|
|
|
In: Eur Phys J C Part Fields (2020)
|
|
BASE
|
|
Show details
|
|
8 |
The role of the precuneus and posterior cingulate cortex in the neural routes to action
|
|
|
|
BASE
|
|
Show details
|
|
9 |
Jointly Learning Entity and Relation Representations for Entity Alignment
|
|
|
|
BASE
|
|
Show details
|
|
10 |
Demographic inference and representative population estimates from multilingual social media data
|
|
|
|
BASE
|
|
Show details
|
|
11 |
Synthetically Supervised Feature Learning for Scene Text Recognition
|
|
Liu, Yang; Wang, Z; Jin, H; Wassell, Ian. - : Springer International Publishing, 2018. : https://doi.org/10.1007/978-3-030-01228-1, 2018. : Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2018
|
|
Abstract:
We address the problem of image feature learning for scene text recognition. The image features in the state-of-the-art methods are learned from large-scale synthetic image datasets. However, most meth- ods only rely on outputs of the synthetic data generation process, namely realistically looking images, and completely ignore the rest of the process. We propose to leverage the parameters that lead to the output images to improve image feature learning. Specifically, for every image out of the data generation process, we obtain the associated parameters and render another “clean” image that is free of select distortion factors that are ap- plied to the output image. Because of the absence of distortion factors, the clean image tends to be easier to recognize than the original image which can serve as supervision. We design a multi-task network with an encoder-discriminator-generator architecture to guide the feature of the original image toward that of the clean image. The experiments show that our method significantly outperforms the state-of-the-art methods on standard scene text recognition benchmarks in the lexicon-free cate- gory. Furthermore, we show that without explicit handling, our method works on challenging cases where input images contain severe geometric distortion, such as text on a curved path.
|
|
URL: https://doi.org/10.17863/CAM.33700 https://www.repository.cam.ac.uk/handle/1810/286387
|
|
BASE
|
|
Hide details
|
|
12 |
Search for dark matter produced in association with heavy-flavor quark pairs in proton-proton collisions at [Formula: see text]
|
|
|
|
BASE
|
|
Show details
|
|
13 |
Altered White Matter Integrity in Adolescents with Prelingual Deafness: A High-Resolution Tract-Based Spatial Statistics Imaging Study
|
|
|
|
In: AJNR Am J Neuroradiol (2013)
|
|
BASE
|
|
Show details
|
|
14 |
Using a Dynamic Neural Field Model to Explore a Direct Collicular Inhibition Account of Inhibition of Return
|
|
|
|
BASE
|
|
Show details
|
|
15 |
The development of a fuzzy expert system to help top decision makers in political and investment domains
|
|
Alshayji, Sameera. - : Brunel University, School of Information Systems, Computing and Mathematics, 2012
|
|
BASE
|
|
Show details
|
|
20 |
Kernel Regression Framework for Machine Translation: UCL System Description for WMT 2008 Shared Translation Task
|
|
|
|
In: Presented at: The Third ACL Workshop on Statistical Machine Translation. (2008) (2008)
|
|
BASE
|
|
Show details
|
|
|
|