AI's Artistic Journey: Capturing Scenes, Missing the Mood with Stability-ai-ultra
- 9 minutes read - 1729 words
Table of Contents
The world of AI image generation is rapidly evolving, with models capable of creating stunning visuals from text prompts. However, achieving a perfect balance between technical accuracy and artistic expression remains a challenge. This blog post delves into the results of an experiment where an AI model was tasked with generating images based on specific scene descriptions, highlighting its strengths and weaknesses in capturing the desired aesthetic.
Created with: stability-ai-ultra
Leap of Joy in the Desert Sun
A woman embraces the freedom of the desert, soaring through the air with a joyful spirit. The bright blue sky and shining sun amplify the sense of adventure and optimism, while the vast sand dunes create a breathtaking backdrop.
Prompt
poses jumping: Excitement, freedom ; A lone adventurer; wide shot; Adventure; a vast, sun-drenched desert landscape; cinematic
Characteristic
Shot : A woman is jumping in the air with her arms spread out in a desert landscape. The sun is shining brightly in the sky and there are sand dunes in the background.
Aesthetic Score : 0.7
Mood : joyful, adventurous, free
Quality
Entropy : 6.47
Noise : 72
Prompt Clip Score : 0.30
AI Evaluation
Likelihood of AI : 0.20
Image errors : There are some artifacts in the sky around the sun, and the sand in the foreground appears a bit grainy.
Silhouetted Joy: A Leap of Hope Against the Sunset City
A person, arms outstretched in triumph, leaps against the backdrop of a vibrant city skyline at sunset. The silhouette against the fiery sky creates a powerful sense of drama and scale, capturing a moment of pure joy and hope.
Prompt
poses jumping: Triumphant, powerful ; A superhero; close-up; Heroism; a cityscape with towering skyscrapers; cinematic
Characteristic
Shot : A person jumps with arms outstretched in front of a cityscape at sunset. The person is silhouetted against the setting sun.
Aesthetic Score : 0.7
Mood : joyful, triumphant, hopeful
Quality
Entropy : 6.86
Noise : 81
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.50
Image errors : The cityscape in the background appears slightly blurry and unrealistic.
Friends Leap for Joy in Mountain Meadow
Capture the spirit of adventure with this vibrant image of five friends jumping in the air against a stunning mountain backdrop. Their infectious smiles and the high-angle perspective create a sense of carefree joy and excitement. The wildflowers and clear blue sky add to the overall feeling of optimism and wonder.
Prompt
poses jumping: Joyful, carefree ; A group of friends; medium shot; Tourism; a scenic mountain vista with a breathtaking view; cinematic
Characteristic
Shot : A group of five young adults, four women and one man, are jumping in the air with their arms raised. They are positioned in front of a mountain range with a blue sky above. There is a green grassy hill in the foreground with yellow wildflowers.
Aesthetic Score : 0.7
Mood : happy, joyful, carefree
Quality
Entropy : 6.94
Noise : 86
Prompt Clip Score : 0.31
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image has some minor blurriness in the background, but it is not overly distracting. The colors are slightly muted, which could be improved with some post-processing.
Mario’s City Leap: A Nostalgic Journey Through Pixelated Streets
This vibrant cartoon captures the essence of classic video game nostalgia, showcasing Mario’s iconic leap through a bustling cityscape. The dynamic pose and playful colors evoke a sense of energy and excitement, reminding us of the joy and adventure found in pixelated worlds.
Prompt
poses jumping: Energetic, playful ; A video game character; close-up; Gaming; a vibrant, pixelated world; cinematic
Characteristic
Shot : Mario in a pixelated city, jumping over buildings, with a bright colorful backdrop.
Aesthetic Score : 0.7
Mood : fun, playful, energetic
Quality
Entropy : 6.58
Noise : 51
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image has some pixelation and aliasing artifacts. The edges of the character’s body are slightly jagged. The background is also lacking detail.
Joyful Leap in a Sea of Travelers
A woman embraces the freedom of flight, leaping into the air in a crowded airport terminal. The sun shines brightly behind her, creating a dramatic contrast against the silhouettes of the bustling crowd. Her outstretched arms and joyful expression capture the liberating spirit of the moment.
Prompt
poses jumping: Anticipation, excitement ; A traveler; long shot; Travel; a bustling airport terminal with people rushing around; cinematic
Characteristic
Shot : A woman jumps in the air with her arms raised in a crowded airport terminal. The sun is shining through the windows and creating a warm glow. The other people in the airport are walking past in a hurry.
Aesthetic Score : 0.7
Mood : happy, excited, joyful
Quality
Entropy : 6.77
Noise : 71
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.10
Image errors : No notable artifacts or errors.
Five Dancers Defy Gravity in a Vibrant Stage Performance
Capture the energy and excitement of a live performance as five female dancers soar through the air, illuminated by colorful stage lights and cheered on by an enthusiastic crowd. The dancers’ mid-air poses create a sense of dynamic movement and exhilaration, perfectly capturing the lively and vibrant mood of the event.
Prompt
poses jumping: Energetic, vibrant ; A group of dancers; medium shot; Groups; a brightly lit stage with a cheering audience; cinematic
Characteristic
Shot : A group of five dancers are performing on stage in front of a cheering crowd. The stage is lit with bright, colorful lights, and the dancers are silhouetted against the lights.
Aesthetic Score : 0.8
Mood : energetic, vibrant, dynamic
Quality
Entropy : 6.72
Noise : 75
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image has some minor noise in the background.
Running from the Storm: A Silhouette of Fear
A dramatic image captures the silhouette of a person fleeing a lightning strike, creating a sense of urgency and danger against a stormy sky. The ominous mood is heightened by the striking contrast between the figure and the lightning.
Prompt
poses jumping: Determined, courageous ; A lone figure; close-up; Heroism; a dark, stormy night with lightning flashing; cinematic
Characteristic
Shot : A person is running away from a lightning strike in a dark, stormy sky. The scene is set in a field with a fence in the background.
Aesthetic Score : 0.6
Mood : dramatic, suspenseful, ominous
Quality
Entropy : 6.94
Noise : 71
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.70
Image errors : The lightning strike is slightly pixelated and the person’s silhouette is not very detailed.
Uncharted Paths Lead to Ancient Wonders
A group of explorers ventures deep into a lush jungle, their path leading towards mysterious ruins in the distance. The vibrant foliage and clear blue sky create a sense of wonder and excitement, hinting at the adventures that await.
Prompt
poses jumping: Curious, adventurous ; A group of explorers; wide shot; Adventure; a dense jungle with ancient ruins; cinematic
Characteristic
Shot : A group of people are walking through a jungle towards a ruined temple.
Aesthetic Score : 0.7
Mood : adventurous, mysterious, hopeful
Quality
Entropy : 6.60
Noise : 99
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.90
Image errors : There are some minor artifacts in the image, particularly in the foliage. The characters also appear to be slightly distorted.
Leap of Faith: A Man Suspended in Time
A dramatic scene unfolds as a man in a dark jacket defies gravity, leaping through a dimly lit room bathed in red and blue light. The action-packed moment is captured with a cool, cinematic flair, highlighting the dramatic effect of the lighting, pose, and sense of motion.
Prompt
poses jumping: Focused, intense ; A gamer; close-up; Gaming; a dimly lit room with a computer screen glowing; cinematic
Characteristic
Shot : A young man in a leather jacket is jumping in a dark room lit with blue and red lights.
Aesthetic Score : 0.7
Mood : dramatic, cool, action
Quality
Entropy : 6.49
Noise : 67
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.40
Image errors : The image appears to be over-saturated. There is a slight blur in the background, but it is not distracting.
Silhouettes of Love Against a Sunset Sky
A couple, bathed in the golden glow of a setting sun, leaps with joy on a sandy beach. Their silhouettes against the vibrant sky create a romantic and playful scene, capturing the essence of a happy and intimate moment.
Prompt
poses jumping: Romantic, carefree ; A couple; medium shot; Travel; a romantic sunset over a beach; cinematic
Characteristic
Shot : A couple is jumping in the air, holding hands, on a beach at sunset. The background is a golden sky and a blue ocean.
Aesthetic Score : 0.7
Mood : romantic, happy, carefree
Quality
Entropy : 6.83
Noise : 72
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.20
Image errors : No obvious errors.
Conclusion
The results show that the generative AI model performed well in understanding the scene and camera position, but struggled with the aesthetic aspect. Here’s a breakdown:
- Camera Position: The model scored 0.45, which is below the “good” range of 0.5 to 0.75. This suggests that the model didn’t perfectly capture the intended camera position described in the prompt.
- Shot Analysis: The model scored 0.63, falling within the “good” range. 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 0.08, which is significantly lower than the “very good” range of -0.2 to 0.1. This suggests that the generated image’s aesthetic deviated from the expected aesthetic described in the prompt.
Overall, the model shows promise in understanding scene descriptions and camera positions, but needs improvement in generating images that match the desired aesthetic.