AI's Eye for Composition: A Mixed Bag of Results with Midjourney
- 10 minutes read - 2022 words
Table of Contents
In the realm of AI-generated imagery, capturing the essence of a scene goes beyond simply placing objects in the right spot. It’s about understanding the nuances of camera positions, shot composition, and the overall aesthetic that brings a scene to life. This blog post delves into an experiment that tested the capabilities of a generative AI model in this regard, revealing both promising results and areas for improvement.
Created with: midjourney
Silhouetted Against the Sunset, a Moment of Reflection
A lone figure stands on the edge of a crumbling stone wall, their silhouette stark against the fiery hues of the setting sun. The vast, hazy landscape stretches out before them, evoking a sense of melancholy and contemplation. Yet, amidst the solitude, a glimmer of hope shines through, suggesting a journey of self-discovery and renewal.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: epic, hopeful ; A lone figure, silhouetted against the setting sun, stands atop a crumbling castle wall; medium shot; heroism; a vast, desolate landscape; cinematic
Characteristic
Shot : A lone figure stands on a stone wall overlooking a vast, golden landscape at sunset.
Aesthetic Score : 0.6
Mood : melancholy, contemplative, hopeful
Quality
Entropy : 6.35
Noise : 102
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.10
Image errors : There are some artifacts and noise visible, particularly in the sky, suggesting possible over-processing or compression.
Lost in the Shadows: A Journey Through the Unknown
Four explorers, cloaked in darkness and armed with flickering torches, navigate a narrow, claustrophobic cave. The air is thick with mystery and suspense as they venture deeper into the unknown, their every step echoing in the silence.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: suspenseful, adventurous ; A group of explorers, their faces illuminated by flickering torchlight, navigate a dark, winding cave; medium shot; adventure; ancient rock formations and dripping water; cinematic
Characteristic
Shot : A group of people are exploring a dark cave, lit only by their lanterns. The cave walls are rough and uneven, and the atmosphere is mysterious and slightly ominous.
Aesthetic Score : 0.6
Mood : dark, mysterious, adventurous
Quality
Entropy : 5.67
Noise : 98
Prompt Clip Score : 0.31
AI Evaluation
Likelihood of AI : 0.10
Image errors : There are slight artifacts and noise in the image, most notably in the darker areas.
Lost in the Neon Maze: A Cyberpunk Gamer’s World
A mid-shot captures the intensity of a gamer immersed in a futuristic city, the neon lights reflecting in their eyes. The dark room and the vibrant game world create a stark contrast, pulling the viewer into the heart of the action.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: intense, focused ; A gamer’s hands, illuminated by the glow of a monitor, deftly manipulate a controller; medium shot; gaming; a vibrant, futuristic cityscape displayed on the screen; cinematic
Characteristic
Shot : A person is playing a video game with a futuristic cityscape in the background. The scene is lit with neon lights.
Aesthetic Score : 0.7
Mood : futuristic, cyberpunk, urban
Quality
Entropy : 6.52
Noise : 102
Prompt Clip Score : 0.35
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image has some slight blurriness and artifacts, especially in the cityscape. The reflection in the glass surface is somewhat odd and inconsistent.
A Family’s Moment of Awe on a Mountaintop
A mid-shot captures a family of three standing on a mountain peak, their figures dwarfed by the vastness of the surrounding landscape. Snow-capped mountains rise in the distance, while a field of wildflowers blooms in the foreground. The scene evokes a sense of serenity, awe, and adventure, highlighting the beauty and power of nature.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: joyful, awe-inspiring ; A family, their faces filled with wonder, stand before a majestic mountain range; medium shot; tourism; a clear blue sky and lush green meadows; cinematic
Characteristic
Shot : A father and his two sons stand in a field, looking out at a snow-capped mountain range in the distance.
Aesthetic Score : 0.8
Mood : peaceful, serene, contemplative
Quality
Entropy : 6.82
Noise : 113
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.70
Image errors : The image appears slightly over-saturated, with the colors appearing slightly artificial. There are also some slight artifacts in the sky and mountains, which could be due to compression or digital manipulation.
Silhouetted Against the Sunset, a Moment of Quiet Reflection
A young woman stands on a rooftop, her silhouette stark against the fiery hues of the setting sun. The cityscape stretches out before her, a canvas of twinkling lights and distant shadows. The scene evokes a sense of serene contemplation, tinged with a hint of wistful loneliness.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: reflective, nostalgic ; A backpacker, gazing out at a breathtaking sunset over a foreign city; medium shot; travel; bustling streets and colorful buildings in the distance; cinematic
Characteristic
Shot : A woman with a backpack stands on a rooftop overlooking a cityscape at sunset.
Aesthetic Score : 0.6
Mood : serene, contemplative, hopeful
Quality
Entropy : 6.61
Noise : 93
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.20
Image errors : There is some noise in the image, particularly in the shadows. There is also a slight color shift in the sky that makes it look unnatural.
A Moment of Wonder
A young girl with curly hair sits on the floor, her eyes wide with curiosity as she gazes upwards. The soft, warm light in the background casts a gentle glow on her face, highlighting her innocent expression. The teddy bear clutched in her hand adds to the sense of childhood wonder and hope.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: anticipatory, heartwarming ; A young girl, her eyes wide with excitement, holds a stuffed animal as she watches her family pack for a road trip; medium shot; family; a cluttered living room filled with suitcases and boxes; cinematic
Characteristic
Shot : A young girl is sitting on the floor in a room that looks like it is being moved. There are boxes and furniture in the background.
Aesthetic Score : 0.7
Mood : sweet, hopeful, curious
Quality
Entropy : 6.80
Noise : 92
Prompt Clip Score : 0.35
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image has some minor artifacts, especially in the shadows and highlights. The girl’s hair looks a little bit blurry.
Heroic Rescue: Firefighter Saves Child from Burning Building
A dramatic image captures the bravery of a firefighter rescuing a child from a burning building. The firefighter’s protective suit and helmet stand in stark contrast to the child’s vulnerability, creating a powerful and emotional scene.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: intense, heroic ; A firefighter, his face grimy with soot, carries a rescued child through the smoke-filled ruins of a building; medium shot; heroism; a burning building in the background; cinematic
Characteristic
Shot : A firefighter is holding a young child in his arms, presumably rescued from a fire. The scene is set against a background of smoke and debris, suggesting a fire scene.
Aesthetic Score : 0.8
Mood : serious, protective, hopeful
Quality
Entropy : 6.68
Noise : 98
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.00
Image errors : There are no significant artifacts or errors in the image.
Starry Night Gatherings: Friends, Fire, and Wonder
A group of friends huddle around a crackling campfire, their silhouettes stark against the breathtaking backdrop of a starry night. The Milky Way stretches across the sky, creating a sense of awe and wonder. The scene is serene, cozy, and contemplative, capturing the magic of shared moments under the vast expanse of the cosmos.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: relaxed, intimate ; A group of friends, their faces lit by the campfire, share stories and laughter under a star-filled sky; medium shot; adventure; a dense forest surrounding the campsite; cinematic
Characteristic
Shot : A group of friends are sitting around a campfire under a starry night sky. The Milky Way is visible in the background.
Aesthetic Score : 0.7
Mood : nostalgic, warm, friendly
Quality
Entropy : 6.03
Noise : 110
Prompt Clip Score : 0.31
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image appears to have been edited to enhance the star field and some artifacts are visible in the sky. The lighting on the faces of the people in the image is inconsistent, and the image is overall a bit dark. The stars are too defined and unnatural.
The Glow of Victory: A Gamer’s Moment of Triumph
A young gamer, bathed in the ethereal light of his screen, captures the raw emotion of victory. The dimly lit room and vibrant background create a dramatic contrast, highlighting the intensity and energy of his triumph.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: exuberant, triumphant ; A gamer, his eyes glued to the screen, celebrates a victory with a triumphant fist pump; medium shot; gaming; a brightly lit gaming room with multiple monitors; cinematic
Characteristic
Shot : A young man is celebrating a victory while playing a video game. He is wearing headphones and has his arms raised in the air. There is a computer monitor behind him, lit up with colorful lights. The photo is taken from a low angle, emphasizing the man’s excitement.
Aesthetic Score : 0.6
Mood : excitement, triumph, joy
Quality
Entropy : 6.44
Noise : 85
Prompt Clip Score : 0.32
AI Evaluation
Likelihood of AI : 0.20
Image errors : Slight blurriness, slight noise, and a slight color cast.
Love Story in the City of Dreams
A couple strolls hand-in-hand down a charming cobblestone street, their silhouettes framed against the backdrop of historic European buildings. The scene evokes a sense of romance, nostalgia, and timeless beauty.
Prompt
Mid-shot or medium-shot Mid-shot or medium-shot: romantic, nostalgic ; A couple, hand in hand, walks along a cobblestone street in a charming European city; medium shot; tourism; quaint shops and cafes lining the street; cinematic
Characteristic
Shot : A couple walking away from the camera down a cobblestone street in a European city. The street is lined with old buildings and there are trees and plants growing on the sides.
Aesthetic Score : 0.7
Mood : romantic, quaint, nostalgic
Quality
Entropy : 6.92
Noise : 105
Prompt Clip Score : 0.36
AI Evaluation
Likelihood of AI : 0.10
Image errors : No major errors, there is a bit of blurriness in the background but it’s not too distracting.
Conclusion
The results show that the generative AI model performed well in understanding and implementing camera positions and shot composition, but struggled with achieving the desired aesthetic. Here’s a breakdown:
Camera Position:
- Score: 0.545
- Interpretation: This score falls within the “good” range (0.5 to 0.75), indicating that the model generally captured the intended camera positions in the generated images. However, it wasn’t quite able to achieve the “very good” level of accuracy.
Shot Analysis:
- Score: 0.645
- Interpretation: Similar to camera position, the shot analysis score falls within the “good” range. This suggests the model understood the scene and its elements well enough to create shots that were generally consistent with the prompt.
Aesthetic Analysis:
- Score: 0.11
- Interpretation: This score is significantly lower than the ideal range of -0.2 to 0.1. This indicates that the generated images didn’t quite match the expected aesthetic style. The model may have struggled with capturing the desired mood, color palette, or overall visual style.
Overall:
The model demonstrates a good understanding of camera positions and shot composition, but needs improvement in achieving the desired aesthetic. This suggests that the model might be better at understanding the technical aspects of image creation than the artistic ones.