method: Tencent TEG OCR2020-03-15
Authors: Pei Xu, Shan Huang, Hongzhen Wang, Shen Huang, Qi Ju.
Description: We reimplemented the standalone recognition method according to the end-to-end text spotting code released by the Mask TextSpotter[TPAMI]. It is a seq-to-seq method based on 2D attention. We synthesize curved text images for pretraining by the method of VGG synthtext. We add public dataset including icdar2013-2015, CUTE, SVT, IIIT5k, RCTW2017, LSVT to finetune and don't use any private data.
method: AntMaskText32023-05-06
Authors: JIe Wang, Tao Huang, Taoxu
Affiliation: Ant-Group
Description: Based on Maskrcnn + SPN + SAM, we select on MaskTextSpotterv3 as our baseline. we construct a dictionary with 5800 classes. Some public datasets are used in our experiments, including icdar2013, icdar2015, LSVT, RCTW, part of MLT, and ReCTS. In the 2D-attention module, we improve the representation ability to recognize the chinese characters.
Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting
method: pq_dp2023-05-05
Authors: -
Affiliation: -
Description: -
Date | Method | Recall | Precision | Hmean | 1-NED | |||
---|---|---|---|---|---|---|---|---|
2020-03-15 | Tencent TEG OCR | 60.33% | 63.66% | 61.95% | 64.13% | |||
2023-05-06 | AntMaskText3 | 52.04% | 64.69% | 57.68% | 59.47% | |||
2023-05-05 | pq_dp | 54.42% | 67.83% | 60.39% | 55.02% | |||
2019-04-28 | baseline_0.5_class_5435 | 47.98% | 52.56% | 50.17% | 54.91% | |||
2019-04-28 | Alibaba-PAI | 47.02% | 62.01% | 53.48% | 51.68% | |||
2019-04-30 | QAQ3 | 41.72% | 55.07% | 47.48% | 49.10% | |||
2019-04-29 | CLTDR | 40.75% | 51.90% | 45.65% | 48.78% | |||
2019-04-30 | Detection-Recognition | 39.71% | 55.02% | 46.13% | 48.03% | |||
2019-11-07 | Sogou_OCR | 42.55% | 47.67% | 44.96% | 46.79% | |||
2021-04-28 | NN_euro6 | 38.46% | 63.53% | 47.91% | 43.89% | |||
2019-04-30 | So Cold 2.0 | 36.25% | 32.26% | 34.14% | 39.58% | |||
2019-04-29 | task3 | 31.98% | 48.62% | 38.58% | 37.65% | |||
2021-04-23 | HOCRA | 37.14% | 40.26% | 38.64% | 37.05% | |||
2021-04-30 | HOCRA_v2 | 35.01% | 53.25% | 42.25% | 36.87% | |||
2023-05-05 | pq_ts | 31.65% | 50.93% | 39.04% | 34.02% | |||
2019-04-16 | Art-test_baseline_task3 | 26.75% | 30.86% | 28.66% | 33.07% | |||
2019-04-30 | CRAFT + TPS-ResNet v1 | 28.12% | 37.82% | 32.26% | 29.58% | |||
2021-04-29 | HOCRA_base | 24.59% | 60.68% | 35.00% | 28.77% |