Codesota · Models · TextBlockV2 (GPT-2)Jiahao Lyu et al., Fudan University1 results · 1 benchmarks
Model card

TextBlockV2 (GPT-2).

Jiahao Lyu et al., Fudan Universityopen-sourceUnknown paramsGPT-2 LM decoder for detection-free scene text spotting

TextBlockV2 uses GPT-2 as the language model decoder for scene text spotting without precise text block detection. arxiv:2403.10047, Mar 2024.

§ 01 · Benchmarks

Every benchmark TextBlockV2 (GPT-2) has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01icdar2015Computer Vision · Optical Character Recognitionaccuracy97.7%#2/292024-03-15source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area

Where TextBlockV2 (GPT-2) actually performs.

Computer Vision
1
benchmark
avg rank #2.0
§ 03 · Papers

1 paper with results for TextBlockV2 (GPT-2).

  1. 2024-03-15· Computer Vision· 1 result

    TextBlockV2: Towards Precise-Detection-Free Scene Text Spotting with Pre-trained Language Model

§ 05 · Sources & freshness

Where these numbers come from.

arxiv
1
result
1 of 1 rows marked verified.