🎨 Artist by day 🌙 Creative Coder by night Computational Artist circa 2016 | PromptEngineering | LLMs | OpenSource | Linux | Computational Muralism
NorCal
Joined January 2016
- Tweets3.1K
- Following369
- Followers718
- Likes10.7K
Visual Synthesizer - Daniel Joy Grimes retweeted
if you want to teach a robot how to draw and paint with both hands at the same time-hit me up. i could even teach it advanced oil painting glazing techniques with my AI/ML skills.
Visual Synthesizer - Daniel Joy Grimes retweeted
Lol, I trained models on my entire collection of decades of handmade art way back in 2019. I don’t always use those models and datasets because it’s fun to explore and synthesize new styles. You can say I’m not a very good artist, but you can’t say I’m not an artist. I am an artist—because I say so.
Visual Synthesizer - Daniel Joy Grimes retweeted
I use AI to generate concepts for commercial storyboards that air on television and have created models for fashion designers, luxury hotels, and other fine artists. My work, under another brand, has been featured in prestigious museums and on products like snowboards. In the hands of a skilled artist, AI can greatly accelerate the concept art process—I can now do the work of 20 artists, and it's possible to create highly detailed murals in just a few hours.
The biggest challenge is that tech companies often don’t hire actual artists, so the tools they release tend to cater more to beginners than to professional artists. Fortunately, I have coding skills and have been building my own AI tools, staying ahead of the curve by about a decade. Eventually, AI will become more useful in commercial art, even for those without technical expertise.
Visual Synthesizer - Daniel Joy Grimes retweeted
From Deep Learning with Python:
- Deep learning research is an evolution process driven by poorly-understood empirical results
- Math in DL papers is usually worthless and was placed there purely as a signal of seriousness
- The key to good research is understanding what *causes* your results, and that requires ablation studies (which few researchers attempt to do)
Visual Synthesizer - Daniel Joy Grimes retweeted
Anyone have a XY LoRA comfyUI testing workflow? My old ones for SDXL using efficiency nodes are not updated yet.
Flux.1 DEV Abstract Surrealism
Visual Synthesizer - Daniel Joy Grimes retweeted
Conceptually there has been no innovation in visual AI art since pix2pix and text2image - all we see is higher resolutions, finer details and more consistency. Change my mind.
Visual Synthesizer - Daniel Joy Grimes retweeted
Sharing how I think about making organized and legible workflows in ComfyUI.
Feat: big labels, groups, color-coding, and Set/Get nodes!
Link below in thread 🧵
Visual Synthesizer - Daniel Joy Grimes retweeted
The progress of the Animatediff community over the past 10 months has been miraculous - see attached!
Now, closed startups like Krea are taking the fruits of all this effort - so I'd like to tell the story of how we got here & what people who believe in open source can do.
Visual Synthesizer - Daniel Joy Grimes retweeted
I'm excited to reveal my most recent project, entirely created using AI, from images, video, sounds, and music. 🎉🤖
Visual Synthesizer - Daniel Joy Grimes retweeted
The dark forest is blossoming at the edges of the internet. Its weapons are encryption, eternal youth and springtime. It has happened before and will happen again.
Do you understand?
Visual Synthesizer - Daniel Joy Grimes retweeted
The skills needed for developing new ML techniques have little overlap with the skills needed for applying ML effectively. A bit like how chip design has little overlap with software engineering.
Visual Synthesizer - Daniel Joy Grimes retweeted
"Reflection & Gratitude" Happy Solstice! Today, I'm dedicating time to reflect on the past year. It's been the most successful year of my professional career, & I'm proud of my accomplishments. For the coming year, I plan to focus on building a vibrant & enjoyable community
