Despite all the work that experts have done on artificial intelligence image generators, they are not able to accurately reproduce the human hand. This issue came to the fore after a Twitter user’s image collection emerged, created by the Midjourney AI creator.
On the one hand, the images turned out very realistic. On the other hand, someone has three hands, someone has seven fingers and a very long palm, etc.
So why can’t artificial intelligence depict a simple object like hands as it should? As Professor Peter Bentley, a computer scientist at University College London, points out, 2D renderers have no idea about the 3D geometry of anything like a hand. So they get the general idea, but none of these models have any idea what it is.
AI systems don’t have any context for creating the right hands in an image. They also cannot take into account the context of the image and have no idea how to do it. In fact, the scientist says, they simply combine everything they have.
However, gradually neural networks learn three-dimensional geometry. This will allow them to create more complex images. Therefore, this may still change in the future. It can also lead to the creation of highly detailed 3D objects and even digital worlds.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
