#科技头条#【AI怎么画出梵高风格的名画?人工智能“学习的进程”从未停止】

#科技头条#【AI怎么画出梵高风格的名画?人工智能“学习的进程”从未停止】继谷歌人工智能可以根据静态图片“脑补”出小视频、绘制画风直追蒙克达利毕加索的梦境图像后,AI文森特可以将简单线条续成“艺术作品”。基于上千幅画作的识别,训练有素的文森特可以熟练运用对比度、配色和结构知识,轻松帮用户完成画作,瞬间令画作水平完成从少儿简笔画到梵高的提升,惊不惊喜,意不意外?

http://www.dailymail.co.uk/sciencetech/article-4928842/AI-computer-transforms-sketches-works-art.html

Move aside, Van Gogh! AI robot 'Vincent' transforms your basic sketches into 'works of art'

  • 'Vincent' uses deep learning to transform sketches into 'works of art'
  • It was created by showing the computer thousands of paintings 
  • This trained Vincent on the use of contrast, colour and texture
  • Users can draw a doodle on a tablet, and Vincent quickly interprets different lines being drawn and picks up where the user left off 

 

If you enjoy art but most of your drawings resemble a child's doodles, then you'll be happy to hear that help is at hand – in the form of an artificial intelligence computer.

Scientists have created a new system called 'Vincent' that uses deep learning to transform sketches into 'works of art.'

Completed 'works of art' combine a users' sketch with art since the renaissance, as if Van Gogh, Cézanne and Picasso were inside the machine.

Scroll down for video

Pictured is one of the sketches
Vincent uses deep learning to transform sketches
SLIDE ME

If you enjoy art but most of your drawings resemble a child's doodles, then you'll be happy to hear that help is at hand – in the form of an artificial intelligence computer. Vincent is the first system with the ability to interpret what a human is drawing, and then complete the piece

WHAT IS VINCENT?

Vincent is the first system with the ability to interpret what a human is drawing, and then complete the piece for them.

To design Vincent, the researchers showed the computer thousands of paintings from the Renaissance period to the current day, training the computer on contrast, colour and texture.

Now it is ready, Vincent can interpret the edges in paintings, and uses this understanding to produce a complete picture.

To use the system, users simply draw directly onto a tablet.

Vincent can then interpret different lines being drawn and pick up where the user left off, taking the information provided to build the piece into a completed picture.

Vincent has been designed by scientists at Cambridge Consultants, and builds on human input to create its masterpieces.

Machine learning has previously been used in the arts to create images and songs based on sounds or pictures it has seen before.

But Vincent is the first system with the ability to interpret what a human is drawing, and then complete the piece for them.

To design Vincent, the researchers showed the computer thousands of paintings from the Renaissance period to the current day, training the computer on contrast, colour and texture.

Now it is ready, Vincent can interpret the edges in paintings, and uses this understanding to produce a complete picture.

Mr Monty Barlow, director of machine learning at Cambridge Consultants, said: 'What we've built would have been unthinkable to the original deep learning pioneers.

 

To use the system, users simply draw directly onto a tablet. Vincent can then interpret different lines being drawn and pick up where the user left off, taking the information provided to build the piece into a completed picture

WHAT IS DEEP LEARNING?

Deep-learning software tries to mimic the activity in layers of neurons in the neocortex, which makes up 80 percent of the brain and is where thinking occurs.

The software learns to recognize patterns in digital representations of sounds, images, and other data.

The idea that software can simulate the neocortex’s neurons in an artificial 'neural network' is decades old, and it has led to disappointments as breakthroughs.

But because of improvements in mathematical formulas and increasingly powerful computers, computer scientists can now model many more layers of virtual neurons than ever before.

Source: MIT Technology Review

'By successfully combining different machine learning approaches, such as adversarial training, perceptual loss, and end-to-end training of stacked networks, we've created something hugely interactive, taking the germ of a sketched idea and allowing the history of human art to run with it.'

To use the system, users simply draw directly onto a tablet.

Vincent can then interpret different lines being drawn and pick up where the user left off, taking the information provided to build the piece into a completed picture.

Beyond art, the researchers believe that there is a range of potential applications for Vincent-like technology.

For example, it could be used in autonomous vehicles if it was trained using driving scenarios and simulations, according to the researchers.

Mr Barlow added: 'We're exploring completely uncharted territory – much of what makes Vincent tick was not known to the machine learning community just a year ago.

'We're excited to be at the leading edge of an emerging, transformative industry and to be making the leap from the art of the possible to delivering practical machine learning solutions for our clients.'

Scientists have created a new system called 'Vincent' that uses deep learning to transform sketches into 'works of art'

Scientists have created a new system called 'Vincent' that uses deep learning to transform sketches into 'works of art'

 


Comments are closed.



无觅相关文章插件