Table Detection and Extraction Using Deep Learning
1–10 of 49 posts
Re: Table Detection and Extraction Using Deep Learning
#2Table extraction has been a feature of better OCR programs for at least a decade. It's easier than the OCR part. Look up "OCR table" for examples, products, code, papers, etc.
Re: Table Detection and Extraction Using Deep Learning
#3Table extraction has been a feature of better OCR programs for at least a decade. It's easier than the OCR part. Look up "OCR table" for examples, products, code, papers, etc.
You'd think that until you try them with tables that contain empty cells that you still need recognized or tables that span multiple pages. I wouldn't say this has been solved for a decade.
Re: Table Detection and Extraction Using Deep Learning
#4Table extraction has been a feature of better OCR programs for at least a decade. It's easier than the OCR part. Look up "OCR table" for examples, products, code, papers, etc.
I wish it was, but it isn't. There are various kinds of tables, that may have delimited lines or not, or they may be unaligned cells, each showing a key and a value... If you actually have in mind some solution that works well (either a paper, a github project, a commercial product) I'd be eager to know
Re: Table Detection and Extraction Using Deep Learning
#5Partially related - is this what someone could use to detect a sudoku grid? The spaces and the digits from a picture?
Re: Table Detection and Extraction Using Deep Learning
#6Table extraction has been a feature of better OCR programs for at least a decade. It's easier than the OCR part. Look up "OCR table" for examples, products, code, papers, etc.
You're wrong.Robust and easy to use table extraction might be solvable, but from a business perspective it isn't solved.
Re: Table Detection and Extraction Using Deep Learning
#7I've worked with several companies that try to parse things in PDF documents, extracting tables and paragraphs etc. This is actually challenging because a PDF is a large bag of words and fragments of words with x y positions. There is a particularly popular word processor that emits individual characters. Just determining that two fragments are part of the same word is challenging as is detecting bullet points, etc.
The AI approaches are definitely still worse than human-written rules. I can infer - and I've chatted with the devs to confirm - from the quality of the text and table extraction whether the company is using a modern NN approach or someone has sat down and handwritten some simple rules that understand indents and baselines etc.
Re: Table Detection and Extraction Using Deep Learning
#8Partially related - is this what someone could use to detect a sudoku grid? The spaces and the digits from a picture?
Sudoku is a relatively simpler problem since the structure is known apriori and becomes as simple as pattern matching.
Re: Table Detection and Extraction Using Deep Learning
#9Partially related - is this what someone could use to detect a sudoku grid? The spaces and the digits from a picture?
Sudoku is a relatively simpler problem since the structure is known apriori and becomes as simple as pattern matching.
Exactly, sudoku can be solved with classical CV through OpenCV, see for example https://www.youtube.com/watch?v=QR66rMS_ZfA
He's using a CNN for digit recognition.
Re: Table Detection and Extraction Using Deep Learning
#10I've worked with several companies that try to parse things in PDF documents, extracting tables and paragraphs etc. This is actually challenging because a PDF is a large bag of words and fragments of words with x y positions. There is a particularly popular word processor that emits individual characters. Just determining that two fragments are part of the same word is challenging as is detecting bullet points, etc.…
I had to check we hadn't worked for the same company! Yeah, text extraction and layout analysis from PDFs is a super interesting challenge and still relatively underdeveloped. I'd put table detection at about the hardest challenge in that field.
One of the contributors to the PDF library I'm developing has been implementing some interesting algorithms for layout analysis https://github.com/UglyToad/PdfPig/wiki/Document-Layout-Anal...