{"id":10402,"date":"2023-08-10T15:41:26","date_gmt":"2023-08-10T13:41:26","guid":{"rendered":"https:\/\/blog.main-vision.com\/?p=10402"},"modified":"2023-08-10T15:49:13","modified_gmt":"2023-08-10T13:49:13","slug":"generating-images-with-bing-ai","status":"publish","type":"post","link":"https:\/\/www.main-vision.com\/richard\/blog\/generating-images-with-bing-ai\/","title":{"rendered":"Generating Images with Bing AI"},"content":{"rendered":"Reading Time: <\/span> 2<\/span> minutes<\/span><\/span>

Yesterday I played with Bing Chat, which is Microsoft’s AI engine and I noticed that I could play with generating images. I spent quite a bit of time generating a multitude of images, in part for fun, but also to get a grasp of the limitations of the opportunity presented by software like AI.<\/p>\n

Faces<\/h2>\n

If you ask Dall-E via Bing Chat to generate a face then it can. It wanted to generate the face of a woman with curly hair so I did, and the image looks realistic. I then asked it to draw the face of a tired man and it did. It’s when I asked Dall-E to draw a woman cyclist exploring the Dolomites that I saw that faces are a mess. It’s fine with generating human faces, in isolation, but if you ask for a landscape, and a human with a face then the face is wrong. This is both great, and a shame. It’s great because it means that we still need to photograph real people doing real things, but it’s bad because if all we need is a face then AI is ready to provide us with some.<\/p>\n

Cycling Ducks<\/h2>\n

I asked Dall-E to create images of ducks cycling around the Vall\u00e9e De Joux and it drew ducks, where instead of legs, the ducks had wheels. Sometimes these wheels were not round. In one case a duck had the front forks and a wheel coming out of the front of its body. Dall-E obviously hasn’t spent much time reading Duck Tales stories. If it had then it would draw flawless ducks.<\/p>\n

Drawing Monuments<\/h2>\n

I asked Dall-E to draw the Palais Wilson with the Ballerina in one instance and it failed. It took a generic building and showed it within the scene I described for it to draw. I also asked it to draw the broken chair at Place Des Nations in Geneva and it failed to do this too. Dall-E doesn’t imagine real things we describe as accurately as I would expect. In fact it doesn’t, at all.<\/p>\n

Firefish<\/h2>\n

I asked Dall-E to draw Firefish providing IT support, staring at a phone, and writing a post. In most of these cases Dall-E worked well. It was able to provide amusing or interesting images. When I asked it to draw a frog waving at a heron it generated one image that was better than the others.<\/p>\n

People and Meerkats Cycling<\/h2>\n

I asked Dall-E to draw people cycling down a cycling route with a railway tunnel and the result was quite good. I then asked for the people to be replaced by meerkats and the result was amusing. If you use your imagination Dall-E can generate some interesting results. <\/p>\n

A Flock of Geese and a camera operator<\/h2>\n

I asked Dall-E to draw a camera operator on a crane filming a flock of geese bobsledding and it drew the camera and the crane well, as well as the geese but it ignored the bobsledding part, and the face of the camera operator was awful.<\/p>\n

And Finally<\/h2>\n

Rather than write more, I will include a series of images on my wordpress blog for you to enjoy.