Posts

*Research: Midjourney + Visily's Screenshot-to-UI + ScreenAI + Adobe Experience Cloud Gen AI + Gpt Sentiment Analysis + Gpt Wordcloud

Image
I like exploring new design methods and tools. I did a quick exploration of Visily's screenshot-to-UI tool. Created a website UI image in Midjourney (6 alpha web version, style = 0, with an aspect ratio of 9:32) out of curiosity to see just how close to something usable and useful you can get, and really to see if it's time saving if you are someone who already creates these. Midjourney for UI Comp I think ideas can come from anywhere, especially visual ideas. Midjourney is another useful tool for roughly exploring an interface idea visually. Currently, the final UI output is still up to you to create, since Midjourney just outputs images. It's getting much much higher fidelity with each update, but I don't think it's loaded up with all the knowledge about every nuance of modern interface design trained into its prompt language just yet. But, it can quickly spit out roughs, so I did that. Side Note: Google's ScreenAI Google's ScreenAI project ( https://resea
Post a Comment
Read more

* Research: Quick exploration of Visily AI's sketch to UI + Figma WireGen AI plugin

Image
I've been exploring different ai text to UI tools lately to see how far they have come with ai. Today's exploration is VisilyAI. https://app.visily.ai/ Ran a quick test of it's sketch to UI design tool. I could have hand drawn a sketch but I decided to use a Figma Plugin called WireGen - AI GPT wireframe generation to draw the sketch. WireGen - AI GPT wireframe generation https://www.figma.com/community/plugin/1221144015267698736/wiregen-ai-gpt-wireframe-generation What I did: Figma > page > open plugins > select WireGen > gave it a prompt. The prompt was something basic to test the functionality only like: "A mobile app to capture ai related content such as links, code, text, images, svgs." Then I download a png of the wireframe sketch frame from Figma. And went into Visily AI and uploaded the UI wireframe png. and it generated the UI design... Then I copied the UI design... and pasted it back into Figma using the Figma Visily AI plugin. Results/Takea
Post a Comment
Read more

Google ScreenAI Sounds Interesting

This is timely. I spent the day trying to see if it’s possible/what sort of UI elements I could fine tune a style model to generate consistently. I've tested a lot of different ai tools to see how they deal with rendering UI elements. That's definitely another, much deeper post. But, this gives another clue as to where ai is with that. Hey Gemini, what is ScreenAI by Google? https://research.google/blog/screenai-a-visual-language-model-for-ui-and-visually-situated-language-understanding/?m=1 Gemini: ScreenAI is a new AI model developed by Google Research that can understand the visual elements present in user interfaces (UIs) and infographics [1]. This allows ScreenAI to perform a variety of tasks, including: Answering questions about information displayed in infographics For instance, if you showed ScreenAI an infographic about the population of different countries, it could answer your question "What is the most populous country?" by looking at the data presented i
Read more

AI Style: Sumi-e Minimalist

Image
I created a new AI style based on a curated set of my sumi-e explorations ( https://www.laetro.com/style-gallery/173 ), and I'm really liking the output...  This ai style is available on Laetro https://www.laetro.com/style-gallery/173
Read more

New AI Habits: Personal GPTs + Q and A with AI

Image
This is approximately a 20 minute read according to Gemini and GPT, and I added a few more minutes to the estimate. Introduction (1 min) GPT Habits (2 min) Personal GPTs (2 min) Using GPTs for UI Design (2 min) Limitations of GPTs (2 min) Early Days. Exploration Continues (2 min) Q and A with AI (8 min) New AI Habits: Personal GPTs  Introduction (1 min) As a designer, I've been exploring AI and Machine Learning for a while now, enough to have sort of created some new habits I'm realizing.  One of those habits that has worked its way into my flow is creating personalized expert GPTs to see how far they go, and to see what's possible. It's kind of an ongoing exploration.  GPT Habits (2 min) With the flood of new AI tools coming out all the time, I keep experimenting with different ways to use AI in UX/UI, visual graphic design, vector illustration, among many other things. I keep trying to see what can be useful toward traditional design goals, and weed out new AI tools
Read more