Details
computerworld.com RSS
Hauptkategorie: IT News aus aller Welt
Kategorie: Computer & Internet
Veröffentlicht: Fr, 02.08.2019 17:30
[Disclosure: NVIDIA is a client of the author.]If I were asked to disclose the product coming to market that I’m most excited about, it’s GauGAN, a software offering from NVIDIA. Part of the reason I’m excited is it addresses a critical shortcoming of mine: I can’t draw to save my life. It also represents a huge step toward the future of personal computing by blending AI capability with a human operator (here’s a video that gives you an idea of what I’m talking about). This is the way many of us think AI should be applied: not to replace humans, but to make us more capable and far more productive.What reminded me of this is that at SIGGRAPH this year, NVIDIA won both best real time live demo and people’s choice awards for GauGAN. Going further, this same approach could disrupt and massively accelerate productivity at a scale heretofore unimagined.GauGANWhat makes GauGAN amazing is that it basically uses an AI to transform people that couldn’t draw to save their lives into world class artists. With a powerful AI back end, the application can take ill-defined drawings and automatically, and almost magically, transform them into photorealistic images.Now this doesn’t just work for us artistically limited folks (personally annoying, because both my mother and stepmother were strong artists) but it also works for artists to significantly speed their efforts. This technology is likely most powerful for animation where each cell often must be individually drawn. GauGAN doesn’t just introduce an object, it integrates that object into the whole picture. The result looks like it was professional drawn, regardless of your skill. For me, it was somewhat like a disabled person suddenly losing their disability: I could suddenly do what I’ve always wanted, but been unable, to do.I look forward to a day when I’ll have access to the product and be able to create the amazing worlds that I can imagine, but just haven’t had the skill to transfer to media so I can share my imagination with the world (or at least a few close friends and my wife). When I saw this, it was almost life changing. I think the only thing more powerful for me would be something that allowed me to fly without a plane and that, too, is on my list of wants. This concept of AI assisting a human operator is what I think will redefine personal computing.Computer aided humanWith GauGAN the operator knows what they want to create and GauGAN makes it happen bridging the user’s skill set with what is needed for them to create photorealistic art. But that same concept is being applied far more broadly. For instance, Chase is using an AI to take rough ad copy and turn it into advertising gold. This is better than using an AI to create derivative work because it still involves the human. We also have programs that look at stories and then automatically create related derivative interactive stories from them.And MIT has created a robot that not only can create its own horror stories but can modify the stories in real time based on tweets from fans. Imagine reading a story, tweeting what you don’t like, and getting a fixed book on your electronic reader nearly instantly. Or, as an author, writing the outline of a series, having the AI complete the books for you and then allow you to tweet your edits and changes. The level of potential AI/human collaboration is unprecedented. Programs like this are already in use: Reporters are using three programs to speed-publish stories that revolve around critical real time events.These are all preferable to just having the AI do it independently, which is also a thing.In the end, the future of the personal computer will likely revolve around this AI capability where the human conceives of the item, the AI refines it, and the human fine tunes the offering to best match his or her imagination, and the AI edits the result to assure people will want to read it.This kind of man-machine collaboration represents the future of computing.I almost cried when I first saw NVIDIA’s GauGAN, because I thought I’d finally be able to draw what I’ve imagined for years. But it’s really just the tip of the iceberg. Future AI-based applications will collaborate, complete, enhance and make our creative efforts far more popular as a stepping stone to our path towards the singularity. As these products mature, they will enhance our skills, improve our capabilities and allow us to do things we’ve always wanted to but never developed – or were unable to develop – the skills for.This blended AI model is the future of personal computing, and products like NVIDIA’s GauGAN are blazing the trail. This article is published as part of the IDG Contributor Network. Want to Join? Copyright © 2019 IDG Communications, Inc.
Erstellt: Fr, 02.08.2019 17:30
aktualisiert: So, 04.08.2019 10:20
[Disclosure: NVIDIA is a client of the author.]If I were asked to disclose the product coming to market that I’m most excited about, it’s GauGAN, a software offering from NVIDIA. Part of the reason I’m excited is it addresses a critical shortcoming of mine: I can’t draw to save my life. It also represents a huge step toward the future of personal computing by blending AI capability with a human operator (here’s a video that gives you an idea of what I’m talking about). This is the way many of us think AI should be applied: not to replace humans, but to make us more capable and far more productive.What reminded me of this is that at SIGGRAPH this year, NVIDIA won both best real time live demo and people’s choice awards for GauGAN. Going further, this same approach could disrupt and massively accelerate productivity at a scale heretofore unimagined.GauGANWhat makes GauGAN amazing is that it basically uses an AI to transform people that couldn’t draw to save their lives into world class artists. With a powerful AI back end, the application can take ill-defined drawings and automatically, and almost magically, transform them into photorealistic images.Now this doesn’t just work for us artistically limited folks (personally annoying, because both my mother and stepmother were strong artists) but it also works for artists to significantly speed their efforts. This technology is likely most powerful for animation where each cell often must be individually drawn. GauGAN doesn’t just introduce an object, it integrates that object into the whole picture. The result looks like it was professional drawn, regardless of your skill. For me, it was somewhat like a disabled person suddenly losing their disability: I could suddenly do what I’ve always wanted, but been unable, to do.I look forward to a day when I’ll have access to the product and be able to create the amazing worlds that I can imagine, but just haven’t had the skill to transfer to media so I can share my imagination with the world (or at least a few close friends and my wife). When I saw this, it was almost life changing. I think the only thing more powerful for me would be something that allowed me to fly without a plane and that, too, is on my list of wants. This concept of AI assisting a human operator is what I think will redefine personal computing.Computer aided humanWith GauGAN the operator knows what they want to create and GauGAN makes it happen bridging the user’s skill set with what is needed for them to create photorealistic art. But that same concept is being applied far more broadly. For instance, Chase is using an AI to take rough ad copy and turn it into advertising gold. This is better than using an AI to create derivative work because it still involves the human. We also have programs that look at stories and then automatically create related derivative interactive stories from them.And MIT has created a robot that not only can create its own horror stories but can modify the stories in real time based on tweets from fans. Imagine reading a story, tweeting what you don’t like, and getting a fixed book on your electronic reader nearly instantly. Or, as an author, writing the outline of a series, having the AI complete the books for you and then allow you to tweet your edits and changes. The level of potential AI/human collaboration is unprecedented. Programs like this are already in use: Reporters are using three programs to speed-publish stories that revolve around critical real time events.These are all preferable to just having the AI do it independently, which is also a thing.In the end, the future of the personal computer will likely revolve around this AI capability where the human conceives of the item, the AI refines it, and the human fine tunes the offering to best match his or her imagination, and the AI edits the result to assure people will want to read it.This kind of man-machine collaboration represents the future of computing.I almost cried when I first saw NVIDIA’s GauGAN, because I thought I’d finally be able to draw what I’ve imagined for years. But it’s really just the tip of the iceberg. Future AI-based applications will collaborate, complete, enhance and make our creative efforts far more popular as a stepping stone to our path towards the singularity. As these products mature, they will enhance our skills, improve our capabilities and allow us to do things we’ve always wanted to but never developed – or were unable to develop – the skills for.This blended AI model is the future of personal computing, and products like NVIDIA’s GauGAN are blazing the trail. This article is published as part of the IDG Contributor Network. Want to Join? Copyright © 2019 IDG Communications, Inc.