Vision FX 2.0 is a brand-new AI art plugin, which will take your imagery and create stunning variations. Perfect for creatives from photographers to digital artists to get the most out of their own work (for example, conceptualising or ideation), the plugin works with a range of creative software – and has plenty of unique selling points.
We got hands-on with the plugin, exploring some of its key features and generation tools – you can learn more about its performance in the video below. And be sure to check out the rest of our AI week content, which is sponsored by Distinct AI, the creators of Vision FX 2.0.
Vision FX 2.0 is a unique first-of-its-kind Gen AI Plugin, for more details and to try it out for yourself, head to Distinct Plugins.io.
01. Use Vision FX 2.0 in a host application
Vision FX 2.0 is a plugin, so you’ll need a host application in order to run it – currently, it’s compatible with CorelDRAW, Corel PHOTO-PAINT, Corel PaintShop Pro, Corel Painter, Adobe Photoshop, and Affinity Photo. Open and access the plugin within the host application of your choice (we used Affinity) and use your own image as a reference to get started.
Note: Make sure that your GPU and drivers are up to date, and install necessary updates beforehand.
Vision FX 2.0 is a local plugin so you’ll need to ensure that your hardware is up to scratch to handle performance-heavy work. Distinct AI recommends using a device equipped with a modern Nvidia graphics card (GeForce 10 series or newer) with a minimum of 12GB VRAM, or a modern Intel integrated or discrete graphics card (Iris Xe or newer).
We found that it’s best to run the plugin on GPU power rather than with a CPU (this can be easily adjusted in the settings menu) for smoother performance, and you should also start with a lower resolution of 640 when experimenting with image creation, and then bump up the output quality to 1080 once you’re ready for a high-res final image to be rendered ready for export.
02. Use your own imagery, and set your prompt
Upload your reference images to get started with Vision FX 2.0. Our resident photographer used images from her own cosplay portfolio to try out Vision FX, and chose a comic book illustration style for the output artwork. The prompt: “Spider-Punk comic book cover in the style of Marvel’s into the spiderverse illustration style, vibrant colours.”
There’s also the option to enter a negative prompt, which can be a set of buzzwords telling the AI what to avoid. For example, you could use “dark hair” or “tall grass” as a negative prompt suggestion if you want the AI to dismiss these features.
Top tip: Be as specific as you can with your prompt if you have a strong idea of what you want, incorporating styles and colour palettes – and you could even include emotions, actions, specific camera lenses you wish to emulate and focal lengths too.
03. Use the sliders to adjust the look
Beneath the prompt box, you’ll find two sets of sliders on the right-hand side which can determine how much the AI generator will allow the output image to stray from the original, and how literally it will take your text prompt.
The image created will reflect the intensity and balance of these choices so it will take some experimentation to discover what works. More strength equals less source material, and an increased text guidance scale should take more of your text prompt into consideration.
04. Refresh new iterations
Underneath those first two sliders, you have the Seed feature, which can only be described as the development of a different artistic output. You can think of artists themselves as seeds in this regard, as no two individuals would paint an identical interpretation.
05. Export up to 4K image quality
Once you’ve made it through the experimentation process and have resulted in a successful image, it’s now time to ram up the quality as high as it will go. This can be done with a few methods. Firstly, increase the step slider towards the bottom as high as it will go (usually 75), this will mean that your image will take longer to generate, however more steps are taken in its creation to ensure quality.
Be sure that the ‘Enhance Image Quality’ toggle button is active, this will tell the AI that you’re after a high quality image. Lastly, you can go to the settings tab in the top toolbar, go to image quality, and set it to 1080p. Again, this will increase the time that your image takes to generate although trust us, it will be worth it.
06. Know your work is secure and private
With Vision FX 2.0, there are no online servers, Discord servers, or other people with access to your projects and creations. This means that along with a no-subscription promise, you have the total freedom to generate whatever you like, whenever you like and re-open your creation back in the original host software that you launched the plugin from. You can then choose to locally save your new image or upload it to a secure cloud storage service of your own choosing.
Head to Distinct Plugins.io to find out more.
Creative Bloq’s AI Week is held in association with DistinctAI, creators of the new plugin Vision FX 2.0, which creates stunning AI art based on your own imagery – perfect for ideation and conceptualising during the creative process. Find out more on the DistinctAI website.
