AI's Artistic Struggle: Capturing the 'style-aesthetic' with Flux-dev
- 9 minutes read - 1799 wordsTable of Contents
The ‘style-aesthetic’ is a captivating artistic style that blends digital and surreal elements, often featuring vibrant colors, distorted perspectives, and a sense of otherworldly wonder. It’s a style that’s difficult to define, yet instantly recognizable. This style is often used in video games, fantasy art, and even in contemporary design. But can AI truly understand and replicate this style? Our recent experiment with a generative AI model reveals both its strengths and limitations in capturing the essence of ‘style-aesthetic’.
Created with: flux-dev
Lost in the Code: A Moment of Intense Focus
A young person, bathed in blue light, stares intently at a computer screen, their headset amplifying their concentration. The blurred background emphasizes their singular focus, creating a sense of drama and suspense. This image captures the raw intensity of dedication and the thrill of the pursuit of knowledge.
Prompt
style-aesthetic Surrealist: Intense and immersive ; A gamer’s face illuminated by the screen; close-up; Gaming; A digital world bleeding into the real world, with characters and objects from the game appearing in the background.; cinematic
Characteristic
Shot : A young person, likely a gamer, is wearing a headset and looking intently at a computer monitor.
Aesthetic Score : 0.7
Mood : focused, intense, contemplative
Quality
Entropy : 6.74
Noise : 80
Prompt Clip Score : 0.22
AI Evaluation
Likelihood of AI : 0.10
Image errors : Slight noise and compression artifacts, especially in the darker areas.
Lost in the City, Found in the Game
A solitary hand grips a black PS4 controller, the city lights blurring into a distant dream. A mood of techy isolation, ready to escape into the digital world.
Prompt
style-aesthetic Surrealist: Intriguing and disorienting ; A gamer’s hand holding a controller; close-up; Gaming; A pixelated world bleeding into the real world, with characters and objects from the game appearing in the background.; cinematic
Characteristic
Shot : A person’s hand holding a black video game controller in a blurred urban background.
Aesthetic Score : 0.6
Mood : tech, gaming, modern
Quality
Entropy : 6.82
Noise : 81
Prompt Clip Score : 0.25
AI Evaluation
Likelihood of AI : 0.20
Image errors : None
A Child’s Wonder in the Forest
A young child, adorned with a vibrant red frog hat, stands amidst a lush forest, dwarfed by a towering orange mushroom. The whimsical scene evokes a sense of childlike curiosity and playfulness, while the dramatic scale of the mushroom creates a feeling of wonder and vulnerability.
Prompt
style-aesthetic Surrealist: Curious and whimsical ; A young adventurer; close-up; Adventure; A jungle filled with giant, talking flowers and glowing mushrooms.; cinematic
Characteristic
Shot : A young child wearing a red frog hat stands in a forest with a giant mushroom behind them.
Aesthetic Score : 0.7
Mood : whimsical, magical, enchanting
Quality
Entropy : 6.82
Noise : 85
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.30
Image errors : The image has some slight blurriness, particularly in the background.
Silhouettes of Serenity: Awe-Inspiring Mountaintop View
Two figures stand on a mountain peak, their silhouettes framed against a breathtaking expanse of clouds and mountains. The soft, hazy light casts a peaceful and contemplative mood, emphasizing the vastness of the scene and creating a sense of awe and wonder.
Prompt
style-aesthetic Surrealist: Romantic and otherworldly ; A couple standing on a mountaintop; long shot; Travel; A mountain range with peaks that reach into the clouds, with a giant, floating city in the distance.; cinematic
Characteristic
Shot : Two figures stand on a mountain peak overlooking a vast, misty mountain range at sunrise, with a prominent, snow-capped peak in the distance.
Aesthetic Score : 0.7
Mood : tranquil, serene, majestic
Quality
Entropy : 6.31
Noise : 55
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.20
Image errors : No noticeable artifacts or errors.
Heroic Silhouette Against the Storm
A powerful superhero stands tall on a skyscraper, their silhouette a beacon of hope against a dramatic backdrop of swirling clouds. The vastness of the cityscape and the stormy sky create a sense of awe and anticipation, hinting at the epic battles to come.
Prompt
style-aesthetic Surrealist: Powerful and unsettling ; A superhero standing on a skyscraper; wide shot; Heroism; A city with buildings that twist and turn like melting wax, with the sky filled with swirling clouds.; cinematic
Characteristic
Shot : A superhero in a red cape stands on top of a tall building, overlooking a city with a dramatic cloudy sky in the background.
Aesthetic Score : 0.6
Mood : epic, dramatic, powerful
Quality
Entropy : 6.63
Noise : 81
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image appears to have been created using AI, with some unnatural edges and textures.
Lost in the Mist: A Solitary Figure Faces Ominous Towers
A lone figure stands shrouded in mist, dwarfed by two towering structures. The clock face on one tower glows ominously, adding to the eerie atmosphere. The scene evokes a sense of mystery, isolation, and impending danger.
Prompt
style-aesthetic Surrealist: Epic and melancholic ; A lone knight; wide shot; Heroism; A vast, surreal landscape with floating castles and giant, melting clocks.; cinematic
Characteristic
Shot : A lone figure stands in a foggy, mysterious landscape, gazing at a towering clock tower and a second, smaller, similar tower in the distance.
Aesthetic Score : 0.7
Mood : eerie, mysterious, atmospheric
Quality
Entropy : 6.55
Noise : 82
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image is slightly blurry, especially in the background, which could be a result of the fog effect. The textures on the towers are a little too smooth and lack detail.
Serene Majesty: Hot Air Balloons Soar Over Majestic Mountains
A breathtaking scene unfolds with a giant hot air balloon dominating the sky, surrounded by smaller companions. The peaceful landscape, dotted with rolling hills and majestic mountains, creates a sense of wonder and tranquility. Three figures stand in the foreground, captivated by the whimsical spectacle above.
Prompt
style-aesthetic Surrealist: Dreamy and fantastical ; A family traveling in a hot air balloon; long shot; Travel; A sky filled with floating islands and giant, whimsical creatures.; cinematic
Characteristic
Shot : A hot air balloon with passengers in the basket flies over a mountain landscape with a couple and a child looking up at it. There are other hot air balloons in the sky and the ground is rocky with trees growing out of the rocks.
Aesthetic Score : 0.7
Mood : peaceful, serene, adventurous
Quality
Entropy : 6.26
Noise : 87
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.30
Image errors : The image is slightly blurry in some areas, particularly the background. Some of the textures, such as the ground and trees, are not very realistic and appear slightly pixelated.
Glowing Flora and Shadowy Figures in a Futuristic Landscape
A mysterious and contemplative scene unfolds in a cavernous blue landscape, where four figures walk amidst glowing vegetation. The composition utilizes light and shadow to create a sense of depth and mystery, leaving viewers with a sense of wonder and intrigue.
Prompt
style-aesthetic Surrealist: Mysterious and awe-inspiring ; A group of adventurers exploring a cave; medium shot; Adventure; A cave filled with glowing crystals and strange, bioluminescent creatures.; cinematic
Characteristic
Shot : Four figures walk through a cave lit by an ethereal blue light.
Aesthetic Score : 0.7
Mood : mysterious, adventurous, hopeful
Quality
Entropy : 6.71
Noise : 95
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.90
Image errors : Some minor aliasing in the shadows and a few pixelated areas.
Pink Paradise: A Selfie with a Sweet Twist
A young woman embraces the playful spirit of a vibrant pink ice cream cone installation, capturing a cheerful selfie that radiates whimsy and joy. The contrast of her pink shirt against the white cones creates a visually striking image, enhanced by warm lighting that adds to the overall sense of cheerfulness.
Prompt
style-aesthetic Surrealist: Humorous and absurd ; A tourist taking a selfie; medium shot; Tourism; A city skyline made entirely of candy, with giant, melting ice cream cones in the background.; cinematic
Characteristic
Shot : A woman in a pink shirt is taking a selfie in a field of giant pink ice cream cones.
Aesthetic Score : 0.6
Mood : playful, summery, whimsical
Quality
Entropy : 6.70
Noise : 93
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image is slightly overexposed, resulting in blown-out highlights on the ice cream cones. The overall color balance is also slightly off, with the colors appearing a bit too saturated.
Peaceful Serenity: Two White Cats Bask in the Sun
Capture the essence of tranquility with this image of two fluffy white cats lounging on a cozy ottoman. The soft lighting and airy window backdrop create a serene atmosphere, while the cats’ relaxed posture adds to the peaceful mood. This image is perfect for evoking feelings of comfort and contentment.
Prompt
style-aesthetic Surrealist: Warm and surreal ; A family portrait; medium shot; Family; A living room with furniture made of clouds and a giant, talking cat.; cinematic
Characteristic
Shot : Two white cats sitting on a stool in a living room with a couch, a large window, and a fake cloud hanging from the ceiling.
Aesthetic Score : 0.7
Mood : cozy, playful, serene
Quality
Entropy : 6.75
Noise : 88
Prompt Clip Score : 0.35
AI Evaluation
Likelihood of AI : 0.20
Image errors : The cloud looks a bit artificial and there is a slight blur in the background.
Conclusion
The results show that the generative AI model performed well in understanding the scene and camera position, but struggled with the aesthetic. Here’s a breakdown:
- Camera Position: The model scored 0.35, which is considered below average. This suggests that the model didn’t accurately capture the intended camera position in the prompt.
- Shot Analysis: The model scored 0.71, which is considered good. This indicates that the model was able to understand the scene described in the prompt and create a shot that aligns with it.
- Aesthetic Analysis: The model scored 0.33, which is considered below average. This means that the generated image’s aesthetic deviated significantly from the expected aesthetic described in the prompt.
Overall, the model demonstrated a good understanding of the scene and shot composition, but struggled to match the desired aesthetic.
Sources:
- https://heartofnoir.com/knowing-noir/aesthetic-of-noir/
- https://www.yellowbrick.co/blog/film/maximizing-the-visual-impact-unveiling-the-art-of-film-aesthetics
- https://www.questjournals.org/jrhss/papers/vol10-issue8/1008255260.pdf
- https://www.jstor.org/stable/3331672
- https://www.cinepoetics.fu-berlin.de/activities/workshops/2020-12-ws/index.html
- https://resource.download.wjec.co.uk/vtc/2016-17/16-17_1-22/eng/Part%201%20What%20is%20Aesthetics.pdf
- https://fal.ai/models/fal-ai/flux/dev/api