Working with AI Images

This has been a busy week for Liz and me. We had to say goodbye to our rented gite in France with its lovely relaxing gardens, patio and heated pool and make our way back to what will probably be cold and wet England. What then could I write about this weekend? Well, as I spend a lot of time making images with AI I thought I could give you a quick account of my experiences, good and bad, with AI imaging.

The idea of AI imaging is pretty simple; on a site like Freepik or Nightcafe, you type into a text box and describe your image and AI or artificial intelligence will do the rest. Here’s a recent example of an image I made for use on social media.

The description was pretty simple and, in my experience, the simpler you can make your description, the better. Once you get complicated, anything can happen. Here it is:

A solitary lighthouse keeper, weathered and wise, stands on a rocky outcrop overlooking a turbulent sea. The lighthouse beam cuts through a thick, mystical fog and the beam projects the following words into the sky: ‘READ A NEW BLOG POST!’ The scene is rendered in a hyper realistic fantasy style. The overall mood is one of stoic endurance and ancient mystery.

The image came out pretty much the way I wanted it. I added the web address later in Microsoft Designer. I also made it into a video which didn’t really work as the lighthouse lamp only turned halfway but the actual light it sent out turned a full circle so it all looked a bit odd. Oh well!

Another idea I had was of a girl hitting a tennis ball and the ball is frozen close to the camera with the words ‘READ A NEW BLOG POST’ seen clearly written on the ball. I must have tried the prompt for this about 30 times and have never really got what I wanted. Here are a few versions.

In the end I thought what about just using the ball without the tennis player? That turned out to be much easier.

Another idea I had was a girl in an art gallery looking at paintings. One would say the usual, A NEW BLOG POST AVAILABLE NOW or something like READ A NEW BLOG POST. The poster would be sharply focussed, the girl blurred but believe it or not every time I tried that prompt, the poster came out blurred and the girl sharp.

Anyway, let’s take a look at some which came out pretty well.

Many of these images I use in my promo videos which can be found in places like Facebook and X. Here’s a recent example.

Freepik is a site where I have always got some good results and recently I spotted an AI voice section on their site. I sorted out quite a few good voices which can be edited and made faster or slower and other refinements but sadly, just as I was getting started, I ran out of credits which was really annoying especially as being a major tightwad I wasn’t inclined to pay for more. Oh well, here’s one voice I used that I thought might resonate with my American readers.

Here’s another video which used an AI image of a bottle floating in the water which came out pretty good as a video. The sound effects were added from a sound effects CD I bought from eBay years ago.

This next one features a pulp fiction paperback with the title, new blog post out now. I tried this prompt a number of times but somehow the text always came out mixed up. This one isn’t perfect; it was supposed to show my name on the spine but shows it on the side where the pages are.

Here’s a final one which will probably reflect the UK weather back in Manchester.

Of course, they don’t call Manchester The Rainy City for nothing!


What to do next: Here are a few options.

Share this post on your favourite social media!

Hit the Subscribe button. Never miss another post!

Listen to my podcast Click here.

Buy the book! Click here to purchase my new poetry collection.

Click here to visit Amazon and download Floating in Space to your Kindle or order the paperback version.

Click here to visit amazon and purchase Timeline, my new anthology.

Graphic Design and More Manipulating the Image

I’m pretty keen on social media, not for social media itself (although it is nice to see what my friends are up to) but as a platform to plug my work, my blog posts and my books and videos. In this post I’m going to look at the impact of graphics and images and the sites I use to create them. All the links open up in a new page.

OK, let’s get started. The first thing to remember about blogs is that a big wall of text tends to put off readers so it’s a good idea to break up the text with a few pictures. What kind of pictures though? Well in the blog posts themselves I’ll use either my own photos or use an online stock photo site like Unsplash.  I’ll also try and render the blog title into a graphic which I’ll also use on Facebook and X (formerly known as Twitter) to promote the post. The easiest way to do that is to use a site like Quotescover where I can just type in the post title and either my name or the name of my website.

The graphic to the right was made on Quotescover.

A site I used to use was Picmonkey which a few years ago was completely free. These days it’s only available to paid users except for the app version which you can use on your phone or, like me, on my iPad. A lot of the features are unavailable to free users though and I use Picmonkey primarily to add text to a photo relevant to my post.

I’ve always liked this picture which I used on a post about my favourite music, The Soundtrack to My Life.

Another site for making graphics or pins for Pinterest is this one, Quozio. Again it’s simple and free to use, just type in a title and your name and choose a template. If you are making a quotation graphic for social media, just add the quotation and whoever said it instead. This one was a simple graphic with a clock.

When Picmonkey declined to let me make free images I had to look elsewhere to make my graphics. One site I began to use was Canva.com. I initially thought Canva was a little complicated but once I got used to it I began to make some good graphics.

Canva has a number of templates which can be personalised and a regular one I use is one for a YouTube thumbnail. What is that you might ask? Well it’s the image you see when you scroll through YouTube looking for an interesting video. A bad thumbnail can turn a viewer off watching the video and good one can pull in a new watcher.

Here are a couple of my YouTube thumbnails. My poetry videos all have a similar thumbnail, just for continuity. Click the image below to watch the video on YouTube.

This one is pretty self explanatory and takes you to my welcome video on YouTube.

Another site I’ve started to use is Adobe Express. It’s similar to Canva and again has various templates that the user can adapt and modify. A graphic I made on Adobe was from a template which I personalised. The picture of me came from Nightcafe, an AI imaging site. How does that work? Well, it creates images from a prompt, so you just describe the image you want to see in words and artificial intelligence does the rest.

How did I get a picture of myself? This is a little trickier. First the user has to upload pictures of the person you want to make an AI model of, in this case me, then you add a prompt to describe the picture you want. On Nightcafe and many other AI sites too, you can upload a picture and develop it. On Freepik you can hit the reimagine button and a different version of your picture will appear.

Sometimes I will make a picture or graphic and send it to another site to see if I can come up with a different or better version.

To make a graphic that I use on X, I added the picture of me to a graphic I had used before and removed the background. Using the animation feature I got the picture to jump into the frame and then jump out again after the text had slid in.

On my laptap I took the animation and added some sound using sound effects downloaded from Zapsplat, a site that provides music and effects for video. Here’s the resulting video.

A lot of the images I’ve created on AI are ones used to plug my blog post over on sites like Facebook and X, graphics that all say something like, ‘New Blog Post Out Now!’

This is one for a post about the Apollo 11 moon landing.

This is for a post about the western genre.

I use three main AI sites, Nightcafe, Freepik and Microsoft Designer. All three sites operate a sort of credit system whereby the user gets so many tokens to create images but after you have used your tokens, that’s it, unless you want to subscribe and pay for more. On Nightcafe, the user gets 5 tokens per day with extra tokens for various things; voting on other users creations, voting on competitions and so on. On Freepik, I’m not quite sure how things work. I’ll sign in and see I have 20 free credits, that’s one credit per picture. Even if I’ve used up my credits I can turn to the ‘reimagine’ page and produce alternative images where I usually have about 5 credits. The next time I visit, sometimes I’ll have another 20 credits, sometimes not, even so, I’ve produced some good images there.

On Microsoft Designer, you can create an image and then use it in a blog graphic or social media post where you can add text and all sorts of stuff.

To create an image you need a prompt. Here’s one I used for an image of a street poster:

A vibrant and eye-catching roadside advertising banner, announcing the release of a “new blog post “ on “www.stevehigginslive.com” The banner features a bold, modern font, with a creative design of a laptop opening up to reveal a digital world. The background is a busy city street, reflecting the urban environment where the blog post was written.

See if you can recognise the result of that prompt further down but one thing to remember is that AI images don’t always turn out the way you want them, especially when using text within the image. Here are a couple that look good but the text didn’t quite work out right.Both of the images above were supposed to say ‘New blog post out now!’ Oh well.

On Nightcafe there is now the option to make an image into a short 4 second video. Here’s one of my favourite promo pictures rendered into a gif, an animated image. The web address was added later on Picmonkey. The image prompt went like this:

Floating in a raging sea during a storm with rain hammering down, we see a bottle on its side with a message inside. The message is visible and says “New blog post out now!” Professional photography, shot on dslr 64 megapixels sharp focus 8k resolution

To finish off, here are a few of my latest graphics, all of which are out there on X in yet another attempt to liven up my posts and to bring in more readers.

OK, what AI image can I create now for my next social media post? What about an art gallery, perhaps seen from above? Light streams in dramatically from windows off to the side. It’s a wide angle shot, looking down. Art lovers are admiring a new poster advertising my blog. It should turn out something like this . .


What to do next: Here are a few options.

Share this post on your favourite social media!

Hit the Subscribe button. Never miss another post!

Listen to my podcast Click here.

Buy the book! Click here to purchase my new poetry collection.

Click here to visit Amazon and download Floating in Space to your Kindle or order the paperback version.

Click here to visit amazon and purchase Timeline, my new anthology.

Manipulating More Images

Once again, it’s time to settle down and write a new blog post. I spend a lot of time putting together a new post not only in terms of the words but equally important are the pictures. I use a lot of  images in my blogs and social media. According to a search I did on Google, social media posts with images get a whopping 94% more views than posts without images. That is a huge statistic and the reason why most internet authors need images. There are plenty of image sites on the internet where bloggers and content creators can source stock pictures but I always tend to try and use my own pictures where I can.

I create a graphic for my blogs every week using various image or graphic sites. The easiest one to use is probably Quotescover.com. It’s a very simple site that combines text and images or just text if you prefer. In Quotescover, you can very simply add your quote or in my case, blog title, add a name or the name of your website and click create. You can choose what type of image you want, for instance for a social media post, for Pinterest, or for a Facebook cover or whatever, then you have a choice of image shape; portrait, landscape or square. After that keep clicking ‘next fonts’ until you get the font that you like.

If I’m feeling a little more creative, I might use Canva which is once again a free site as long as you are making some basic images. At Canva the user can choose from a range of templates and there are various ones available for free but then to use the better ones you have to upgrade to the paid version. You can save your image in various ways, as a simple jpg, or if better quality is required, a png. You can also animate an image or graphic and save it as either a gif, just like the one seen here, which is just an animated picture, or as a short video. By saving your graphic as a video, the user can add music or sound effects, both are available from Canva but of course then you have to update to the paid version.

Most of the animated images I create are used in my blogs or on Twitter as gifs although sometimes I get really creative and add music or sound effects. scroll down to see an example below.

I’ve got a CD of sound effects at home that Liz got me for Christmas ages ago but sometimes I download them from sites like Zapsplat which has a huge library of free sounds.

Another interesting development in imaging is artificial intelligence, more commonly known as AI. We hear a lot about AI in the media lately. Things like what will it mean for all of us. I’m not sure it will mean a great deal unless you use a computer but then again these days, everyone uses a computer, even if it’s only the one in your mobile phone.

A collection of AI pictures of me made using my iPad.

One thing I’ve found interesting about AI is its use in imaging and as I have already said, imaging is vitally important in my blogs and videos. Aside from that I’m interested in images anyway. One of the great things about digital photography is the way images can be altered by editing. Years ago, I’d get a roll of film and have it developed and if there was a dud shot, there wasn’t much I could do about it. I’ve sometimes had almost an entire roll of film turn out to be pretty poor and still had to pay for the developing and printing but fast forward to the 21st century and things are different.

Pictures can be cropped and rotated. Dark pictures can be lightened and bright ones darkened. OK, some people could actually do those things in the past as long as they had access to a great deal of equipment but the great thing today is that anyone with a computer or a tablet can now change the images they produce. Add AI into the mix and pictures can be further transformed. Blurred pictures can be made sharp, backgrounds can be changed and even new images can be created by artificial intelligence.

Take a look at this next photo. You might think for a moment that it’s me, using an old-fashioned typewriter to knock out my next blog post. Come to think of it, that looks like a pretty perfect location to be writing a blog post and I can just imagine myself there, tapping away on a typewriter, which were my exact thoughts when I created it. I didn’t pose for the camera. I used a site called Nightcafe to produce the image. First, I had to create a ‘model’ of myself and to do so I had to upload a lot of photos of me with a minimum of 16 being required. After digesting this visual information Nightcafe created a visual model that I can use in my images.

The images themselves come from a text suggestion. After selecting my model my text went something like this: author typing away on a typewriter with an exciting landscape seen through the French windows.

There are other choices to be made too, choices of style and lighting and so on. I chose a photorealistic style but I could have chosen anime, hyperreal, impressionist, fantasy and many others.

Some of the images I’ve made look like me and some do not. Here’s my favourite which is a pretty good replica of my face.

Some time ago I used an app for my iPad. I eventually deleted it as it seemed to stop working but for £2.99, I was able to create 50 images of myself. Once again, I had to upload a number of photos of me. The results ranged from nothing like me at all to some that actually made me better looking than I really am.

I’ve noticed also that on many of the graphics apps and programs that I use there are options for AI to assist with the text or the backgrounds. These are always extra options and as I always tend to use the free versions of apps like this, I’ve not be able to use them.

Another great image and graphics editor is Adobe Express. Once again, I use the free version but even so, on the free version I’ve made a number of what I think are pretty impressive animated graphics, some of which I always tag onto the end of my YouTube videos.

Here’s another one, also made on adobe with a soundtrack downloaded from Zapsplat. It was mainly used on Twitter (X as they call it now). It was animated from a graphic I used on a blog post called Don’t Make Me Laugh.

Deepfake videos are ones in which the image of someone famous is inserted into another video. According to the internet, recent deepfake videos have been made of people like Barack Obama, Donald Trump and Taylor Swift. What is a deepfake video? Well, it’s a video powered by AI in which images of well-known people are digitally merged into video footage to create something wholly new and fake. Fake sexually explicit images of Taylor Swift were recently distributed on the internet and had many views before they were taken down. X which used be Twitter said recently it was actively removing all identified images and is taking action against the accounts responsible.

AI certainly has great implications for the future, even more realistic images for use on the internet and in the cinema. Better special effects for Star Trek and superhero films and hopefully, even better graphics for blogs like this one. Even the Royal Family have been featured in digital imaging news, did they digitally alter a picture of Kate Middleton and her children? Oh dear.

Anyway, that’s enough blogging for now. Think I might go out for a jog. well, not a real jog, maybe just an AI one!


What to do next: Here are a few options.

Share this post on your favourite social media!

Hit the Subscribe button. Never miss another post!

Listen to my podcast Click here.

Buy the book! Click here to purchase my new poetry anthology.

Click here to visit Amazon and download Floating in Space to your Kindle or order the paperback version.