AI's Artistic Struggle: Capturing the Essence of Dramatic Poses with Stable-diffusion
- 9 minutes read - 1778 wordsTable of Contents
Dramatic poses are a powerful tool in visual storytelling, conveying emotions and narratives through the positioning of the human body. From heroic figures standing against a stormy backdrop to silhouetted adventurers against a setting sun, these poses evoke a sense of drama and intrigue. In this blog post, we explore the challenges of using AI to generate images based on dramatic poses, analyzing the results of a recent experiment and highlighting the model’s strengths and weaknesses.
Created with: stability-ai-core
A Lone Figure Braces Against the Storm
A dramatic scene unfolds as a solitary figure stands on a cliff, facing the fury of a stormy sea. Lightning illuminates the sky, while crashing waves and the figure’s isolated stance create a sense of foreboding and impending danger.
Prompt
poses silhouette: epic, determined ; Lone figure standing on a clifftop, overlooking a vast, stormy sea; wide shot; heroism; dramatic sky with lightning; cinematic
Characteristic
Shot : A lone figure in a coat stands on a cliff overlooking a stormy sea, with lightning striking in the background.
Aesthetic Score : 0.7
Mood : dramatic, ominous, solitary
Quality
Entropy : 6.71
Noise : 65
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.60
Image errors : The image appears to be slightly over-exposed, resulting in some loss of detail in the darker areas.
Silhouettes of Hope Against a Desert Sunset
Five figures stand silhouetted against a breathtaking desert sunset, their forms merging with the vast landscape. The scene evokes a sense of serenity, adventure, and hope, leaving viewers to ponder the stories behind these enigmatic figures.
Prompt
poses silhouette: hopeful, adventurous ; A group of adventurers silhouetted against the setting sun, walking towards a distant mountain range; medium shot; adventure; desert landscape; cinematic
Characteristic
Shot : Five people are walking in a desert landscape during sunset. The sun is setting behind a mountain range in the distance. The people are silhouetted against the sunset.
Aesthetic Score : 0.7
Mood : tranquil, adventurous, hopeful
Quality
Entropy : 6.50
Noise : 53
Prompt Clip Score : 0.32
AI Evaluation
Likelihood of AI : 0.10
Image errors : None
Lost in the Digital Shadows: A Young Man’s Intense Focus
A dimly lit room, a silhouette hunched over a computer screen. The air is thick with intensity as a young man engages in a digital world, his focus unwavering. The mystery of his task and the dramatic lighting create a sense of intrigue, leaving you wondering what secrets lie within the digital realm.
Prompt
poses silhouette: intense, focused ; A gamer’s hands silhouetted against a glowing computer screen, holding a controller; close-up; gaming; neon lights and digital interfaces; cinematic
Characteristic
Shot : A person sitting in front of a computer, typing on a keyboard. The room is dimly lit with blue and red light.
Aesthetic Score : 0.6
Mood : dark, intense, focused
Quality
Entropy : 5.72
Noise : 52
Prompt Clip Score : 0.23
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is a bit noisy, and the colors are slightly oversaturated.
Eiffel Tower Romance: A Silhouette of Love at Sunset
A couple stands silhouetted against the iconic Eiffel Tower at sunset, creating a romantic and dreamy scene. The dramatic effect of their figures against the backdrop evokes a sense of intimacy and mystery, capturing the essence of a nostalgic moment.
Prompt
poses silhouette: romantic, nostalgic ; A couple holding hands, silhouetted against the iconic Eiffel Tower; medium shot; tourism; Parisian cityscape at night; cinematic
Characteristic
Shot : A couple silhouetted against the Eiffel Tower in the distance. The sky is a soft blue and orange color as the sun sets.
Aesthetic Score : 0.7
Mood : romantic, dreamy, nostalgic
Quality
Entropy : 6.57
Noise : 54
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.20
Image errors : No major errors, but the image is slightly overexposed and lacks detail in the shadows.
Silhouetted Against the Setting Sun: A Lone Figure’s Journey
A solitary figure walks a dusty road in a desolate desert landscape, silhouetted against a vibrant sunset. The vastness of the scene evokes feelings of solitude, mystery, and a glimmer of hope. The dramatic use of silhouette and the open landscape creates a sense of isolation and contemplation, inviting viewers to ponder the figure’s journey and the secrets held within the desert.
Prompt
poses silhouette: lonely, contemplative ; A lone traveler walking down a dusty road, silhouetted against the rising sun; long shot; travel; vast, open desert landscape; cinematic
Characteristic
Shot : A lone figure walks down a dirt road in a desert landscape. The sun is setting in the background, casting a warm glow over the scene.
Aesthetic Score : 0.7
Mood : lonely, hopeful, dramatic
Quality
Entropy : 6.76
Noise : 73
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.20
Image errors : There are no significant errors in the image.
Silhouettes of Joy: Friends Celebrate in a Dimly Lit Bar
A group of friends raise their glasses in a dimly lit bar, their silhouettes creating a sense of mystery and intrigue. The scene captures the joyful and celebratory mood of the evening, with a touch of relaxed intimacy.
Prompt
poses silhouette: joyful, celebratory ; A group of friends raising their glasses in a toast, silhouetted against a brightly lit bar; medium shot; groups; vibrant nightlife scene; cinematic
Characteristic
Shot : A group of friends are toasting each other with drinks in a dimly lit bar.
Aesthetic Score : 0.5
Mood : festive, celebratory, joyous
Quality
Entropy : 5.69
Noise : 56
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image has some noise and graininess due to the low light conditions, and there are some slight blurriness around the edges of some subjects.
Batman Takes Flight at Dawn
A dramatic low-angle shot captures Batman soaring above a cityscape at dawn, his silhouette a powerful symbol of heroism against the hazy sky. The scene evokes a sense of adventure and grandeur, promising an epic tale to unfold.
Prompt
poses silhouette: powerful, heroic ; A superhero leaping from a tall building, silhouetted against the city skyline; wide shot; heroism; cityscape with skyscrapers; cinematic
Characteristic
Shot : A superhero, likely Batman, is flying above a cityscape. The cityscape looks like New York City.
Aesthetic Score : 0.6
Mood : dramatic, heroic, mysterious
Quality
Entropy : 6.63
Noise : 71
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.80
Image errors : There are some artifacts and errors in the image, particularly in the cityscape. Some of the buildings look unrealistic and the overall texture of the cityscape is somewhat artificial.
Emerging from the Shadows: Soldiers in a Misty Jungle
A group of five soldiers, silhouetted against a misty jungle landscape, emerge from a dark cave. The dramatic use of light and shadow creates a sense of mystery and suspense, drawing the viewer’s eye to their unknown mission.
Prompt
poses silhouette: suspenseful, adventurous ; A group of explorers silhouetted against the entrance to a dark, mysterious cave; medium shot; adventure; dense jungle foliage; cinematic
Characteristic
Shot : A group of soldiers in silhouette stand at the mouth of a cave in a jungle, looking out towards a foggy, mysterious valley beyond.
Aesthetic Score : 0.6
Mood : mysterious, suspenseful, dramatic
Quality
Entropy : 4.37
Noise : 74
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image has some minor artifacts, particularly in the shadows and around the edges of the objects.
Lost in the Neon Glow: A Hacker’s Silhouette
A shadowy figure hunches over a keyboard, bathed in the ethereal glow of pink and blue neon lights. Two computer monitors flicker with abstract patterns, hinting at a world of secrets and digital intrigue. This mysterious scene evokes a sense of focused intensity, leaving the viewer wondering what secrets lie hidden within the code.
Prompt
poses silhouette: intense, focused ; A gamer’s hands silhouetted against a glowing computer screen, typing furiously; close-up; gaming; futuristic, neon-lit gaming room; cinematic
Characteristic
Shot : A man is sitting in front of a computer, typing on a keyboard, with his back to the viewer. The image is lit by neon lights in the background, giving it a cool and futuristic feel.
Aesthetic Score : 0.7
Mood : cyberpunk, futuristic, mysterious
Quality
Entropy : 5.53
Noise : 50
Prompt Clip Score : 0.22
AI Evaluation
Likelihood of AI : 0.20
Image errors : No visible artifacts, however, the image could be slightly more detailed for a more engaging visual experience.
Silhouettes of Love: A Family’s Sunset Embrace
A tranquil scene of a family of five silhouetted against a vibrant sunset on a beach, framed by swaying palm trees. The image evokes a sense of unity, togetherness, and nostalgia, capturing the beauty of shared moments.
Prompt
poses silhouette: peaceful, heartwarming ; A family standing on a beach, silhouetted against the setting sun; medium shot; tourism; tropical beach with palm trees; cinematic
Characteristic
Shot : A family of four silhouetted against a sunset on a beach, with palm trees in the foreground.
Aesthetic Score : 0.7
Mood : tranquil, peaceful, nostalgic
Quality
Entropy : 5.85
Noise : 61
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.20
Image errors : No significant 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 0.3, which is considered below average. This suggests that the model didn’t accurately capture the intended camera position described in the prompt.
- Shot Analysis: The model scored 0.465, which is also below average. This indicates that the model didn’t fully understand the scene described in the prompt and didn’t create an image that accurately reflects it.
- Aesthetic Analysis: The model scored 0.08, which is considered very good. This means that the generated image closely matched the expected aesthetic style, despite the issues with camera position and scene understanding.
Overall, the model seems to be better at capturing the desired aesthetic than understanding the scene and camera position. This suggests that the model might need further training to improve its ability to interpret and translate prompts into accurate visual representations.