Document Layout Analysis

"Document Layout Analysis is performed to determine physical structure of a document, that is, to determine document components. These document components can consist of single connected components-regions [...] of pixels that are adjacent to form single regions [...] , or group of text lines. A text line is a group of characters, symbols, and words that are adjacent, “relatively close” to each other and through which a straight line can be drawn (usually with horizontal or vertical orientation)." L. O'Gorman, "The document spectrum for page layout analysis," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, no. 11, pp. 1162-1173, Nov. 1993.

Image credit: PubLayNet: largest dataset ever for document layout analysis

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 81–90 of 99 papers

Title	Date	Tasks	Status	Hype
LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding	Dec 29, 2020	Document Image ClassificationDocument Layout Analysis	CodeCode Available	0
Multiple Document Datasets Pre-training Improves Text Line Detection With Deep Neural Networks	Dec 28, 2020	Document Layout AnalysisLine Detection	—Unverified	0
Training data-efficient image transformers & distillation through attention	Dec 23, 2020	Document Image ClassificationDocument Layout Analysis	CodeCode Available	1
docExtractor: An off-the-shelf historical document element extraction	Dec 15, 2020	Document Layout AnalysisSegmentation	CodeCode Available	1
Vision-Based Layout Detection from Scientific Literature using Recurrent Convolutional Neural Networks	Oct 18, 2020	Document Layout Analysisobject-detection	—Unverified	0
VisualWordGrid: Information Extraction From Scanned Documents Using A Multimodal Approach	Oct 5, 2020	Document Layout Analysis	—Unverified	0
CDeC-Net: Composite Deformable Cascade Network for Table Detection in Document Images	Aug 25, 2020	Document Layout AnalysisTable Detection	CodeCode Available	1
DocBank: A Benchmark Dataset for Document Layout Analysis	Jun 1, 2020	Document Layout Analysis	CodeCode Available	1
A Large Dataset of Historical Japanese Documents with Complex Layouts	Apr 18, 2020	Document Layout Analysis	CodeCode Available	3
Combining Visual and Textual Features for Semantic Segmentation of Historical Newspapers	Feb 14, 2020	Document Layout AnalysisSemantic Segmentation	CodeCode Available	1

Show:10 25 50

← PrevPage 9 of 10Next →

All datasets PubLayNet val U-DIADS-Bib D4LA Document Layout Recognition Challenge mini-dev Document Layout Recognition Challenge test RVL-CDIP

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	CDeC-Net	Table	0.98	—	Unverified
2	VGT	Overall	0.96	—	Unverified
3	TRDLU	Overall	0.96	—	Unverified
4	DETR	Overall	0.96	—	Unverified
5	VSR	Overall	0.96	—	Unverified
6	LayoutLMv3-B	Overall	0.95	—	Unverified
7	DiT-L	Overall	0.95	—	Unverified
8	DoPTA	Overall	0.95	—	Unverified
9	UDoc	Overall	0.94	—	Unverified
10	ResNext-101-32×8d	Overall	0.94	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	CV-Group	Class Average IoU	83.4	—	Unverified
2	CNKI	Class Average IoU	77.8	—	Unverified
3	VAI-OCR	Class Average IoU	70.7	—	Unverified
4	DeepLabV3+	Class Average IoU	66.5	—	Unverified
5	L3i++	Class Average IoU (Few-shot setting)	61.1	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DoPTA	mAP	70.72	—	Unverified
2	DocLayout-YOLO	mAP	70.3	—	Unverified
3	VGT	mAP	68.8	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Faster_RCNN	Overall	0.96	—	Unverified
2	fglihai	Overall	0.96	—	Unverified
3	Faster-RCNN	Overall	0.95	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	fglihai	Overall	0.92	—	Unverified
2	USYD NLP_CS29-2	Overall	0.92	—	Unverified
3	Faster-RCNN	Overall	0.91	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	VisualWordGrid	FAR	28.7	—	Unverified