Rapid Prototyping Services, Professional manufacturer of CNC Prototyping and 3D Prototyping in China. 

Recreating Paintings With AI And 3D Printing

by:Tuowei     2019-09-10
There seems to be some contradiction in using artificial intelligence to create art, but the creative work of computers is increasingly considered to be of legitimate value.
In a recent auction by Christie\'s, the world\'s largest and most famous auction house for artificial intelligence
Auxiliary art is sold for $432,500, which makes critics frantically guess the future of AI art and whether it can be considered a work of art.
No matter which side you are on, art artificial intelligence is gaining traction, artists adopt it in many different ways, just like they would use any other tool to create original work or recreate the most valuable work in the world.
Due to the collateral damage or destruction of priceless works of art, famous artists have recently made bold statements about our attitude towards \"price\" art, and computers may --
The generated paintings change our view of art and its value?
Robot artist robot art competition, founded in 2016 by Stanford University graduate Andrew connolu, attracted artists, mechanics and computer scientists interested in advancing the field of robot art.
This year\'s winner, CloudPainter (
Built and programmed by Pindar Van Arman)
Using deep learning, cameras and selected 3D print brush heads, more and more autonomy is achieved in the painting produced by CloudPainter
But as Van Alman said in a HBO & HBO documentary last year, \"my robot . . . . . . Making the same creative decision as I do, which seems to mean that although allowing the project to make a decision based on his instructions, he is still an \"artist \". The third-
CMIT ReArt from the University of Kasestart, Thailand places entries in different ways and accurately reproduces the artist\'s strokes, including the force and position of the brush, creating an exact copy of the drawing strokes through strokes.
This approach illustrates the differences in the current approach to making AI artwork: some artists try to make AI created independently, while others try to reproduce human abilities in form and function.
While \"original\" AI artwork creates a lot of noise and debate in the art world, AI replication becomes more accurate and can solve problems that have plagued the art world since dawn of time.
Reproduction of classical reproductions has always been an important part of the art world, adding value to the original work and democratizing the artist\'s work --
In fact, print and reproductions have recently become the fastest growing part of the art market.
A faithful replica is more than just spreading the artist\'s work.
So close to the original that it is considered a replica of the fax that can be used to replace and preserve the original artwork, or to restore the classic painting lost by fire, theft, or the fog of time.
Historically, it is difficult to create these faxes for many reasons, but new technologies have emerged that address these issues while seeking to make the art reproduction process easier and more accurate.
A group of researchers at mit csail proposed a new way to reproduce painting using deep learning and 3D printing.
The new system \"redraw\" uses a mixture of two methods to precisely recreate the color: a new technology called \"color control\" that uses 10 thin layers of transparent ink, this is an older technique that uses a large amount of ink dots to create continuous colors, similar to the classic dot-color painting technique.
This hybrid technology addresses three key issues in 2D printed replicas: limited spectra (
Only cyan, magenta, yellow and black (CYMK))
, Strict \"total ink limit\", if more than \"will cause the image quality to decline, ink blockage or poor mechanical function\", the 2D printer can create a good reproduction under a lighting condition, and it is not accurate under another lighting condition (
Problem called \"meta mode).
The 3D printer uses color control and halftone, mixes a mixture of ten different inks, creates a series of \"main\" colors that match the target painting, and then places a small ink on the canvas, to reproduce the mixed color or half tone that appears in the painting.
Choosing which transparent inks to use in the \"stack\" is the work of a deep learning program that trains original painting images under different lighting conditions.
In this way, the program learns the \"real\" spectral color of the painting, instead of picking up the distortion of the light in a 2D image, and can predict under any lighting condition, which series of ink will produce the most faithful reproduction.
Although there are still a lot of improvements to redraw before it is commercially feasible (
Certain colors like dark blue cannot be reproduced through their Ink Range)
Zhang Rijin, one of the authors of the paper, said that the system works under any lighting condition, showing a much larger color reproduction capability than almost any other previous work.
Due to the cost of 3D printing, RePaint\'s replica is currently limited to the size of the business card, but the team hopes that commercial printing can mitigate this cost once the project is fully developed.
However, this may offset some of the more selfless consequences of creating art faxes, such as making it easier for those who cannot experience the original art to have access to the art that is important in culture and history.
The innate creativity of artificial intelligence may be an issue of debate, and the question remains whether the art created by artificial intelligence should be viewed like a work that humans have spent years imagining,, there is no doubt that AI is good at using input data to achieve a given output.
When the data is colored and the output is an undistinguished copy of \"lilisa\", does this reduce the value of the Da Vinci original, or add value to the concept of \"Mona Lisa\" by letting more people appreciate its beauty?
Regardless of whether an AI program is considered a \"talent \"(
Therefore, it should be attributed to the value of their work)
Creating faithful reproductions of famous art helps to preserve these paintings for future generations and avoid expensive time
Consumption recovery (
Read: repeated damage to Mona Lisa).
Redrawing may not be ready to deal with the world\'s most precious artwork, but to achieve true AI artists by avoiding competition, the mit csail team may have created the most talented replicas in the world without falling into what \"art talent\" should mean.
Custom message
Chat Online
Chat Online
Leave Your Message inputting...