Vitruvian Man
Da Vinci Series
After Series
- 5The Fall of Man, after Titian, c.1550
- 6Modern Day Myth and Mythology, after Ben Fink, 2023
- 7Wave (SOTT), after Tom of Finland, 1974
- 10The Fall of Man, after Titian, c.1550
The Vitruvian Man is a well-known drawing by Leonardo da Vinci, depicting a male figure in two superimposed positions with arms and legs spread. It symbolizes the ideal proportions of the human body and the relationship between art and science.
Images
I have mentioned previously that Leonardo da Vinci holds a special place in my heart. He has always been a strong influence in my life because his works often combine the fields of art and science — a process and practice that I employ in my own works for the past 30 years.
To be honest, I have always struggled to fully articulate or explain my cross-disciplinary approach to others. It has proven to be more challenging than trying to reason with individuals who hold homophobic beliefs.
Without deviating onto an unrelated tangent, I felt compelled to create AI generative works inspired by the profound impact the Vitruvian Man by Leonardo da Vinci has had on me. That’s what these works represent—a homage to the iconic masterpiece.
This is the first of many to come. I have additionally created some that are much “adherent” to the original, but will likely only post those on my site because of obvious reasons.
Tech
These were done with the Line Art control net. Specifically, I took the image of the Vitruvian Man found on Wikpedia, and converted it into black and white inside Photoshop. I then tweaked the Curves to make the lines more pronounced. Then I wrote my text prompts inside txt2img to get the exact render I need. The more true-to-form subseries (which shall be posted later in the future) has prompts that differ from these series of run in running shorts.
I chose Airfuck’s Brute Mix — Airfuck’s latest model — for its painterly photographic quality.
These were rendered with Euler a, mostly because the DPM++ samplers appeared to have issues in the Automatic1111’s 1.6.x releases, which I later discovered other people had experienced as well. The Euler a sampler had improved in the 1.6.x release, so I decided to use it for its speed and effectiveness when combined with ABM, producing the photorealistic painterly effect I love.
Stable Diffusion txt2img with Line Art control net.
- Airfuck’s Brute Mix v1.0
- Euler a, 20 steps, 512x512
- vae-ft-mse-840000-ema-pruned
- Denoising 0.5, Clip skip 2
- Line Art control net: Weight 1, Start 0.3, End 1, Res 1024, Balanced mode
- Hires: 10 steps, 2x, 1024x1024, 8x_NMKD-Superscale_150000_G
- Automatic1111 v1.6.0
- Post Upscale: Topaz Gigapixel HQ 4x
- Color Correction: Adobe Lightroom
Images
Images
Additional Images
Images
Funny Errors
Images
Full Technical Parameters
- Steps: 20
- Sampler: Euler a
- CFG scale: 4
- Face restoration: CodeFormer
- Size: 512x512
- Model hash: 8ff8347bb0
- Model: airfucksBruteMix_v10
- VAE hash: 235745af8d
- VAE: vae-ft-mse-840000-ema-pruned.vae.ckpt
- Denoising strength: 0.5
- Clip skip: 2
- ControlNet 0:
- Module: invert (from white bg & black line)
- Model: control_v11p_sd15_lineart (43d4be0d)
- Weight: 1
- Resize Mode: Crop and Resize
- Low Vram: False
- Processor Res: 1024
- Guidance Start: 0.3
- Guidance End: 1
- Pixel Perfect: False
- Control Mode: Balanced
- Hires upscale: 2
- Hires steps: 10
- Hires upscaler: 8x_NMKD-Superscale_150000_G
- Lora hashes:
- bulgerk-dickprint: b1934ecdac9a
- epi_noiseoffset2: d1131f7207d6
- epiCRealismHelper: c5e24d9bc4b5
- TI hashes:
- Asian-Less-Neg: 22d2f003e76f
- NegfeetV2: df90b1ff666d
- face_editor_enabled: True
- face_editor_face_margin: 1.6
- face_editor_confidence: 0.97
- face_editor_strength1: 0.3
- face_editor_strength2: 0
- face_editor_max_face_count: 20
- face_editor_mask_size: 0
- face_editor_mask_blur: 12
- face_editor_prompt_for_face: handsome face
- face_editor_apply_inside_mask_only: True
- face_editor_apply_scripts_to_faces: False
- face_editor_face_size: 512
- face_editor_use_minimal_area: False
- face_editor_ignore_larger_faces: True
- face_editor_affected_areas: Face
- face_editor_upscaler: None
- face_editor_tilt_adjustment_threshold: 40
- Version: v1.6.0
- Post Upscale: Topaz Gigapixel HQ 4x
- Color correction: Adobe Lightroom