Slashdot Mirror
Microsoft Excel Can Now Turn Pictures of Tables Into Actual, Editable Tables (thurrott.com)
Microsoft has rolled out a new feature to Excel's Android app that makes it easy to capture data. From a report: Excel now lets you take pictures of a document/paper in real life, crop the picture, and turn that into an actual, editable data on Excel. After capturing the data, you can edit the data to make sure Excel's image recognition is 100% accurate, and make any changes if some of the scanned data were incorrect. The company says it will roll out this feature to Excel for iOS app soon.
.NORM Normal File Format
I totally read that as edible tables.
Good is never enough, when you dream of being great!
You don't know how many times that I received a screen shot of some text in a Word document. But I've gotten the same thing a number of times in an Excel spreadsheet as well. They should put this in desktop Excel.
Well on a PC what would you do? try to hold paper up to your shitty webcam blocking the screen so you can't see what it is capturing or focusing on?
Can the PC version excel do this with PDF files? That should be a lot easier.
Yes, edible table put out by Iodine Bucks the same guy that makes the pepperoni plates.
Why, because traditional tables are stupid!
Easy answer...It is HARD. OCR has been around for years but even today it is hit or miss except for the most sophisticated systems.
I'm curious how well it translates on a mobile device. I've used OCR on PDF and Word documents that were just images and even when the text was perfectly legible it had a hard time getting the letters right and the formatting was always all over the place.
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
So you can convert an image of numbers into data in a spreadsheet? It's a trivial improvement on something we have been able to do for decades.
Wake me up when the software can discover the relationship between the columns in the table, and insert the appropriate cell-references and math operations.
If it weren't for deadlines, nothing would be late.
Easy answer...It is HARD. OCR has been around for years but even today it is hit or miss except for the most sophisticated systems. I'm curious how well it translates on a mobile device. I've used OCR on PDF and Word documents that were just images and even when the text was perfectly legible it had a hard time getting the letters right and the formatting was always all over the place.
Pretty sure the OCR will not be done on your device, but elsewhere.
The Daddy casts sleep on the Baby. The Baby resists!
Because Excel will give you a 2D representation to work with from the cropped picture; to use 3D modeling program, you need to take photos of the table from multiple perspectives, in order to build a 3D model.
Because Microsoft has had an Office suite monopoly for soooo looooong that they haven't needed to actually innovate until now.
Yes, that is my view as well.
... you know, the one where they develop an app that takes a picture of an equation and then the app solves the equation?
How about adding a spreadsheet diff tool that isn't a useless piece of shit designed without usability in mind?
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
It's not that hard. I used Fine Reader to OCR stuff more than a decade ago and it worked well enough on printed text to be very useful. Complex layout with tables, multiple columns of text and so on could be a challenge, but knowing it's supposed to be an Excel table, combined with a decade of neural net development, it should've been relatively trivial.