AI's Struggle with Camera Angles: A Look at Facial Expressions in Generated Images with Scenario
- 10 minutes read - 2030 wordsTable of Contents
Facial expressions are a powerful tool for conveying emotions and storytelling in art. In the realm of AI-generated images, the ability to create realistic and expressive faces is a crucial aspect of achieving a compelling visual experience. This blog post delves into a recent study that examined the performance of generative AI models in capturing facial expressions, specifically focusing on the model’s ability to understand and respond to camera positions. The study revealed that while AI excels at capturing the emotional nuances of facial expressions, it struggles with accurately representing camera angles. This finding highlights the ongoing challenges and opportunities in the development of AI art, particularly in achieving a holistic understanding of visual composition.
Created with: scenario
Lost in the Rain: A Moment of Solitude
A woman walks alone through the rain-soaked streets, illuminated by the soft glow of streetlights. The image evokes a sense of melancholy and contemplation, capturing the feeling of being lost in thought and surrounded by the quiet solitude of the night.
Prompt
facial-expressions Shame: Desolate, lonely, regretful ; A lone figure, hunched over, walking down a deserted street; eye-level; Single Person; Rain-slicked pavement and flickering streetlights; cinematic
Characteristic
Shot : A lone woman walks down a rainy street at night, her silhouette is visible against the light from the streetlamps reflecting in the puddles.
Aesthetic Score : 0.7
Mood : gloomy, melancholic, atmospheric
Quality
Entropy : 6.63
Noise : 113
Prompt Clip Score : 0.25
AI Evaluation
Likelihood of AI : 0.70
Image errors : The image appears slightly blurry, especially around the edges of the buildings. The rain drops are somewhat repetitive and lack variation in size and shape.
Golden Hour Reflections
A young woman stands silhouetted against a vibrant cityscape at sunset, her long brown hair flowing in the wind. The golden light casts a warm glow, evoking a sense of melancholy, thoughtfulness, and a glimmer of hope.
Prompt
facial-expressions Shame: Melancholy, disillusioned, burdened ; A superhero, their mask removed, revealing a face etched with pain; eye-level; Hero; A cityscape bathed in the glow of a setting sun; cinematic
Characteristic
Shot : A woman with long brown hair wearing a white hooded jacket stands in front of a city skyline at sunset. The sunset is a soft orange and pink.
Aesthetic Score : 0.8
Mood : dreamy, hopeful, melancholic
Quality
Entropy : 6.61
Noise : 92
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.90
Image errors : Some slight blurriness in the background and around the hair. Minor artifacts in the hair.
Lost in Thought: A Moment of Melancholy in a Retro Diner
A woman sits alone in a vintage diner, her gaze fixed on something unseen. The soft lighting and intimate composition create a sense of isolation and introspection, inviting viewers to share in her quiet contemplation. The scene evokes a feeling of nostalgia and melancholy, leaving us to wonder what thoughts are swirling in her mind.
Prompt
facial-expressions Shame: Embarrassed, defeated, self-loathing ; A woman, her face buried in her hands, sitting alone at a crowded diner table; eye-level; Normal Person; The bustling activity of the diner, a stark contrast to her isolation; cinematic
Characteristic
Shot : A woman sits at a table in a diner, looking forlornly at a plate of food. Other people are in the background, but the focus is on the woman. The image has a retro feel, like it is from the 1950s or 1960s.
Aesthetic Score : 0.7
Mood : melancholy, lonely, nostalgic
Quality
Entropy : 6.73
Noise : 98
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.60
Image errors : The image is slightly blurry, and the colors are a bit muted.
Immersed in the Game: A Gamer’s Focused Serenity
A young woman, headphones on, sits at her desk with dual monitors, completely engrossed in a video game. The image captures the casual yet focused mood, highlighting the serene determination of a gamer in their element.
Prompt
facial-expressions Shame: Empty, defeated, lost in a digital world ; A gamer, staring blankly at a screen, his controller lying idle; eye-level; Gamer; A dimly lit room filled with gaming paraphernalia, a sense of disconnection; cinematic
Characteristic
Shot : A young woman is sitting in a gaming chair at her desk, facing a dual monitor setup. She has headphones on and is looking towards the right of the frame.
Aesthetic Score : 0.8
Mood : focused, playful, thoughtful
Quality
Entropy : 6.66
Noise : 92
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image appears to have some slight artifacts in the woman’s hair and skin. The lighting seems a bit artificial and the shadows might need a slight adjustment.
Lost in the Crowd: A Man’s Solitary Reflection
A stark black and white pencil sketch captures a man in a suit, standing apart from a bustling crowd. His contemplative expression and isolated position evoke a sense of melancholic introspection and loneliness.
Prompt
facial-expressions Shame: Anxious, self-conscious, out of place ; A man, standing in a crowded room, his eyes darting nervously around; eye-level; Single Person; A party scene, filled with laughter and conversation, but he feels isolated; cinematic
Characteristic
Shot : A black and white drawing of a man standing out from a crowd of people
Aesthetic Score : 0.6
Mood : dramatic, pensive, hopeful
Quality
Entropy : 6.58
Noise : 111
Prompt Clip Score : 0.21
AI Evaluation
Likelihood of AI : 0.50
Image errors : Some of the faces in the crowd are blurry and lack detail. The drawing style is slightly inconsistent.
Silhouettes and Sunset: A Moment of Contemplation
A young man stands alone on a rooftop, bathed in the warm glow of a setting sun. The cityscape stretches out before him, a blur of buildings against the vibrant sky. His solitary figure evokes a sense of melancholy and contemplation, while the vastness of the scene creates a feeling of serenity.
Prompt
facial-expressions Shame: Disheartened, disillusioned, questioning his purpose ; A hero, standing on a rooftop, looking down at the city below; not too close; Hero; A panoramic view of the city, but he feels small and insignificant; cinematic
Characteristic
Shot : A young man stands on a rooftop overlooking a cityscape, possibly New York City, at sunset.
Aesthetic Score : 0.75
Mood : melancholy, contemplative, hopeful
Quality
Entropy : 6.71
Noise : 90
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.80
Image errors : Some slight blurring in the background, especially in the cityscape. The subject’s hair appears slightly unnatural, almost plastic-like.
A Moment of Quiet Contemplation
A woman sits alone in a kitchen, her pensive expression and the simple setting creating a sense of quiet solitude and melancholy. The image evokes a feeling of introspective reflection, capturing a moment of quiet contemplation.
Prompt
facial-expressions Shame: Depressed, unmotivated, lost in her thoughts ; A woman, sitting at her kitchen table, staring at a plate of untouched food; eye-level; Normal Person; A cluttered kitchen, a reflection of her inner turmoil; cinematic
Characteristic
Shot : A young woman in an apron sits at a kitchen table with a salad in front of her. She rests her head on her hand and stares off into the distance. The kitchen is clean and well-lit, with a window behind her that lets in natural light.
Aesthetic Score : 0.8
Mood : melancholy, contemplative, peaceful
Quality
Entropy : 6.66
Noise : 94
Prompt Clip Score : 0.30
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image seems overly smoothed and is slightly blurry. It could be due to the way the image was created or a post processing effect.
The Focused Mind: A Minimalist Portrait of Concentration
A young man sits engrossed in his work, the glow of the computer screen illuminating his face. The minimalist style emphasizes the calm and focused mood, while the dramatic use of light and shadow adds a touch of intensity to the scene.
Prompt
facial-expressions Shame: Despair, addiction, a sense of being lost ; A gamer, hunched over his keyboard, his fingers flying across the keys, but his eyes are filled with sadness; eye-level; Gamer; A brightly lit gaming room, but he feels trapped in a digital world; cinematic
Characteristic
Shot : A young man sitting at a desk in front of a computer, with a minimalist background, a few additional monitors around him, and a slight mess of cables.
Aesthetic Score : 0.7
Mood : calm, focused, introspective
Quality
Entropy : 6.14
Noise : 78
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.80
Image errors : The lines are sometimes a bit jagged, and there is a slight blurring on the background. Some of the details are also somewhat blurry, such as the monitors and the cables.
Lost in Thought: A Moment of Melancholy in Black and White
A solitary figure walks through a bustling city street, his gaze fixed on the pavement below. The stark black and white imagery evokes a sense of loneliness and contemplation, highlighting the man’s introspective mood. The blurred background further emphasizes his isolation, creating a poignant image of quiet reflection amidst the urban chaos.
Prompt
facial-expressions Shame: Rejected, isolated, a sense of being unwanted ; A man, walking away from a group of people, his head down, his shoulders slumped; eye-level; Single Person; A bustling street, but he feels alone and invisible; cinematic
Characteristic
Shot : A young man walks down a city street, with a crowd of people behind him. The image is in black and white.
Aesthetic Score : 0.6
Mood : melancholy, contemplative, urban
Quality
Entropy : 6.70
Noise : 101
Prompt Clip Score : 0.21
AI Evaluation
Likelihood of AI : 0.30
Image errors : The image has a slight amount of blurriness, particularly in the background. The subject’s hair looks a bit unnatural.
A Hero’s Burden: Youthful Knight Amidst Ruins
A young man in silver armor stands defiant before a crumbling building, his face illuminated by a soft light. The scene evokes a sense of melancholy and stoicism, highlighting the dramatic irony of his youthful appearance against the backdrop of destruction. The soft lighting enhances the image’s aesthetic appeal, emphasizing the hero’s features and the weight of his burden.
Prompt
facial-expressions Shame: Guilt, regret, a sense of responsibility ; A hero, standing in the ruins of a battle, his armor dented and his face covered in grime; not too close; Hero; A scene of destruction, a reminder of the cost of his actions; cinematic
Characteristic
Shot : A young man in armor stands in front of a crumbling stone building. The scene is likely set in a medieval or fantasy world.
Aesthetic Score : 0.75
Mood : dramatic, melancholic, heroic
Quality
Entropy : 6.80
Noise : 112
Prompt Clip Score : 0.23
AI Evaluation
Likelihood of AI : 0.30
Image errors : No noticeable artifacts or errors. Some blurriness and lack of detail in the background and other elements.
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 0.2, indicating it’s not very good at reacting to camera positions in the prompt. A score between 0.5 and 0.75 would be considered good, and above 0.75 very good.
- Shot Analysis: The model scored 0.68, which is good at understanding the scene in the prompt. A score between 0.5 and 0.75 is considered good, and above 0.75 very good.
- Aesthetic Analysis: The model scored 0.04, which is very good at matching the expected aesthetic of the image. A score between -0.2 and 0.1 is considered very good.
Overall, the model seems to be better at understanding the scene and achieving the desired aesthetic than it is at reacting to camera positions.
Sources:
- https://dramaresource.com/storytelling/
- https://seedsoftellers.eu/resources/the-body-language-for-young-tellers/
- https://digitalcollections.sit.edu/cgi/viewcontent.cgi?article=1288&context=sandanona&filename=1&type=additional
- https://citeseerx.ist.psu.edu/document?doi=7f842882e9bb1fa2c0e96939bc8d2c37e34e17c0&repid=rep1&type=pdf
- https://www.twinkl.co.uk/search?q=drama+facial+expression
- https://www.scenario.com