玉米棒子
发表于 2025-3-28 15:39:25
SynthTIGER: Synthetic Text Image GEneratoR Towards Better Text Recognition Models the combination of synthetic datasets, MJSynth (MJ) and SynthText (ST). Our ablation study demonstrates the benefits of using sub-components of SynthTIGER and the guideline on generating synthetic text images for STR models. Our implementation is publicly available at ..
Debark
发表于 2025-3-28 19:28:31
http://reply.papertrans.cn/29/2824/282313/282313_42.png
不能约
发表于 2025-3-28 23:17:02
http://reply.papertrans.cn/29/2824/282313/282313_43.png
不在灌木丛中
发表于 2025-3-29 04:05:06
http://reply.papertrans.cn/29/2824/282313/282313_44.png
ingenue
发表于 2025-3-29 07:18:14
http://reply.papertrans.cn/29/2824/282313/282313_45.png
Herbivorous
发表于 2025-3-29 14:08:49
http://reply.papertrans.cn/29/2824/282313/282313_46.png
Liability
发表于 2025-3-29 19:27:55
Fast Text vs. Non-text Classification of Imagess, as encountered in social networks, for detection and recognition of scene text. The proposed classifier efficiently removes non-text images from consideration, thus allowing to apply the potentially computationally heavy scene text detection and OCR on only a fraction of the images..The proposed
harbinger
发表于 2025-3-29 22:33:33
Mask Scene Text Recognizer a supervised learning task of predicting text image mask into a CNN (convolutional neural network)-Transformer framework for scene text recognition. The incorporated mask predicting branch is connected in parallel with the CNN backbone, and the predicted mask is used as attention weights for the fe
激励
发表于 2025-3-30 03:29:46
http://reply.papertrans.cn/29/2824/282313/282313_49.png
kidney
发表于 2025-3-30 06:04:55
Heterogeneous Network Based Semi-supervised Learning for Scene Text Recognitionbased on abundant labeled data for model training. Obtaining text images is a relatively easy process, but labeling them is quite expensive. To alleviate the dependence on labeled data, semi-supervised learning which combines labeled and unlabeled data seems to be a reasonable solution, and is prove