PK PyTorchKR Papers
Papers with Code · CC BY-SA 4.0

Optical Character Recognition (OCR)

89 평가 기록 30 datasets

Optical Character Recognition or Optical Character Reader (OCR) is the electronic or mechanical conversion of images of typed, handwritten or printed text into machine-encoded text, whether from a scanned document, a photo of a document, a scene-photo (for example the text on signs and billboards in a landscape photo, license plates in cars...) or from subtitle text superimposed on an image (for example: from a television broadcast)

개요는 각 항목의 일부입니다. 위 탭에서 전체 기록을 페이지별로 탐색할 수 있습니다.

벤치마크

Papers with Code 아카이브의 데이터셋별 기록 예시입니다. 평가 조건과 지표 방향이 검증되지 않아 순위를 의미하지 않습니다.

Dataset기록된 모델 예시MetricPaperCode
IAM
DTrOCR 105M
CER: 2.38
논문 코드
DTrOCR
Accuracy (%): 89.6
논문 코드
TypiClust
Accuracy: 93.2
논문 코드
HTR-VT
Test CER: 2.8Test WER: 7.4
논문 코드
TrOCR
Test CER: 3.4Test WER: -
논문 코드
HTR-VT
Test CER: 3.9Test WER: 16.5
논문 코드
GPT-4o
Character Error Rate (CER): 0.2378Word Error Rate (WER): 0.5117Average Accuracy: 76.22
논문 코드
PyLaia (all transcriptions + agreement-based split)
CER (%): 4.34WER (%): 15.14
논문 -
Flor
CER: 6.52
논문 코드
AttentionOCR_Inception-resnet-v2_Location
Sequence error: 15.8
논문 코드
BN-DRISHTI Line Segmentation
F-Score: 0.9997
논문 코드
I2L-NOPOOL
BLEU: 89.09%
논문 코드

데이터셋

30개
IAMBenchmarking Chinese Text Recognition: Datasets, Baselines, and an Empirical StudyCIFAR10 (10,000)LAM(line-level)IAM(line-level)READ2016(line-level)VideoDB's OCR Benchmark Public CollectionBelfortKOHTDFSNS - TestBN-HTRdI2L-140K

논문

라이브러리

Papers with Code 리더보드 row에 연결된 구현 저장소입니다.

github.com/video-db/ocr-benchmark 1 논문
github.com/untrix/im2latex 1 논문
github.com/aliiafkari/SUT_Dataset 1 논문
github.com/leandroschelb/attention_ocr 1 논문
github.com/LinearPi/OCR_Chinese 1 논문
github.com/OzHsu23/chineseocr 1 논문
github.com/tensorflow/models 1 논문
github.com/tensorflow/models/tree/master/research/attention_ocr 1 논문

하위 태스크

HT Handwritten Text Recognition 13 데이터셋 · 46 rows AL Active Learning 3 데이터셋 · 10 rows HR Handwriting Recognition 6 데이터셋 · 10 rows HD Handwritten Digit Recognition 2 데이터셋 · 3 rows GD Grapheme Detection 0 데이터셋 · 0 rows HC Handwritten Chinese Text Recognition 0 데이터셋 · 0 rows HD Handwritten Digit Image Synthesis 0 데이터셋 · 0 rows IT Irregular Text Recognition 0 데이터셋 · 0 rows OH Offline Handwritten Chinese Character Recognition 0 데이터셋 · 0 rows WS Word Spotting In Handwritten Documents 0 데이터셋 · 0 rows