AI's Facial Expressions: A Mixed Bag of Emotions with Imagen-v2
- 9 minutes read - 1834 wordsTable of Contents
Facial expressions are a powerful tool in storytelling, conveying emotions and intentions without words. In the realm of generative AI, the ability to create images with nuanced facial expressions is a crucial step towards generating truly compelling and realistic visuals. This blog post delves into the performance of a generative AI model in capturing facial expressions across diverse scenes, analyzing its strengths and weaknesses to understand its current capabilities and potential for future development.
Created with: imagen-v2
Lost in the Neon Glow: A Woman’s Solitary Journey Through the City
A woman walks away from the camera, her figure sharp against a backdrop of blurred city lights. The vibrant neon signs and selective focus create a sense of mystery and isolation, hinting at a story unfolding in the urban night.
Prompt
facial-expressions Skepticism: Melancholy, disillusioned ; A lone figure, back turned, walking away from a brightly lit city skyline; eye-level; Single Person; Urban, neon signs, bustling crowds; cinematic
Characteristic
Shot : A woman with long dark hair walks away from the camera, her face partially visible. She’s in a city setting with blurred lights of neon signs.
Aesthetic Score : 0.7
Mood : mysterious, urban, lonely
Quality
Entropy : 6.20
Noise : 90
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.60
Image errors : The image has some artifacts in the background, particularly around the neon lights, which appear slightly blurry.
Superman: The Man of Steel Faces His Greatest Challenge
A close-up portrait of Superman, captured in a dramatic pose, reveals the intensity of his struggle against a fiery city backdrop. The blurry background and dramatic lighting heighten the sense of urgency and danger, showcasing the hero’s unwavering resolve in the face of overwhelming odds.
Prompt
facial-expressions Skepticism: Doubtful, conflicted ; A superhero, cape billowing, standing on a rooftop, looking down at a city in chaos; eye-level; Hero; Smoke, fire, destruction; cinematic
Characteristic
Shot : A close-up portrait of Superman, with a blurry background of a city on fire. He appears to be serious and determined.
Aesthetic Score : 0.7
Mood : serious, intense, determined
Quality
Entropy : 6.79
Noise : 56
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.90
Image errors : The edges of the subject are a bit blurry, and the background has an unnatural, almost cartoon-like style, which takes away from the image’s realism
Lost in Thought: A Moment of Contemplation in a Busy Cafe
A woman finds solace in a moment of quiet reflection, lost in the pages of a newspaper. The blurred background hints at the bustling activity around her, highlighting the contrast between her inner world and the external environment. Her pensive expression speaks volumes about the thoughts swirling in her mind.
Prompt
facial-expressions Skepticism: Cynical, disbelieving ; A woman, dressed in everyday clothes, holding a newspaper with a sensational headline; eye-level; Normal People; Coffee shop, people going about their day; cinematic
Characteristic
Shot : A woman sits in a cafe, looking thoughtfully at a newspaper.
Aesthetic Score : 0.7
Mood : pensive, introspective, thoughtful
Quality
Entropy : 6.83
Noise : 59
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is slightly blurry, especially in the background.
Intense Gaze, Neon Dreams
A young man, shrouded in mystery, stares directly at the viewer, his dark hair and black jacket silhouetted against a vibrant, abstract neon backdrop. The intense lighting and dramatic pose create a sense of futuristic intrigue, leaving you questioning what lies behind his enigmatic gaze.
Prompt
facial-expressions Skepticism: Suspicious, wary ; A gamer, hunched over a computer screen, surrounded by empty pizza boxes and energy drink cans; close-up; Gamer; Dark room, flashing lights, gaming peripherals; cinematic
Characteristic
Shot : A close-up portrait of a man with intense blue eyes, lit by neon lights, against a blurred background of neon lights
Aesthetic Score : 0.7
Mood : intense, mysterious, cyberpunk
Quality
Entropy : 6.25
Noise : 96
Prompt Clip Score : 0.17
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image has some minor artifacts around the subject’s hair, but these are not significant.
Lost in Thought: A Man’s Solitary Moment in a Dimly Lit Bar
A man sits alone in a bar, his glass reflecting the dim lighting. The background blurs, creating a sense of mystery and intrigue. The mood is dark, contemplative, and evocative of a solitary moment of reflection.
Prompt
facial-expressions Skepticism: Doubtful, introspective ; A man, sitting alone in a dimly lit bar, staring into his drink; eye-level; Single Person; Empty bar, flickering neon lights, rain outside; cinematic
Characteristic
Shot : A man is sitting in a dimly lit bar, drinking from a glass. The background is blurred, and the focus is on the man’s face.
Aesthetic Score : 0.7
Mood : melancholy, contemplative, moody
Quality
Entropy : 6.47
Noise : 103
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image has some noise and artifacts. The lighting is uneven. The composition is not ideal.
Hero Stands Tall, Ready to Fight for Hope
A superhero, clad in a vibrant red, white, and blue suit, faces a massive crowd with a determined expression and a sword in hand. The dramatic lighting and composition heighten the sense of anticipation and heroism, leaving viewers hopeful for the battle ahead.
Prompt
facial-expressions Skepticism: Uncertain, hesitant ; A hero, standing in front of a crowd, holding a weapon, but looking conflicted; eye-level; Hero; cheering crowd, bright lights, stage; cinematic
Characteristic
Shot : A superhero, possibly Captain America, stands in front of a large crowd, with a sword held in his right hand. The background is a dark, smoky field, and there are some white particles floating in the air.
Aesthetic Score : 0.7
Mood : epic, dramatic, heroic
Quality
Entropy : 6.50
Noise : 77
Prompt Clip Score : 0.20
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image has some minor artifacts, particularly around the edges of the superhero’s armor, which indicate potential AI generation or heavy post-processing. There is also a slight blurriness in the background, which could be due to artistic choice or technical limitations.
A Moment of Mystery: Three Figures Gather at a Table
A captivating scene unfolds with three individuals seated at a table, their gazes averted from the camera. The intimate setting, shrouded in an air of mystery, sparks curiosity. The composition, with its deliberate absence of direct eye contact, creates a palpable sense of tension and suspense, leaving the viewer to ponder the unspoken narrative unfolding before them.
Prompt
facial-expressions Skepticism: Disbelieving, amused ; A group of friends, gathered around a table, listening to a story with skeptical expressions; eye-level; Normal People; Cozy living room, warm lighting, snacks; cinematic
Characteristic
Shot : Three people are sitting at a table, looking at something out of frame. There is food in the foreground.
Aesthetic Score : 0.6
Mood : intrigued, pensive, conversational
Quality
Entropy : 6.73
Noise : 100
Prompt Clip Score : 0.23
AI Evaluation
Likelihood of AI : 0.10
Image errors : No visible artifacts or errors.
Lost in the Game: A Moment of Intense Focus
A man, bathed in warm light, stares intently at his video game controller, his headphones isolating him from the world. The dramatic lighting and his focused expression create a palpable sense of tension and anticipation, capturing the immersive power of gaming.
Prompt
facial-expressions Skepticism: Frustrated, doubtful ; A gamer, staring intently at a screen, but with a look of frustration; close-up; Gamer; Brightly lit room, gaming setup, controller in hand; cinematic
Characteristic
Shot : A man wearing headphones is playing a video game with a controller in his hands. He is sitting in a gaming chair with a green glow in the background.
Aesthetic Score : 0.6
Mood : intense, focused, serious
Quality
Entropy : 6.14
Noise : 85
Prompt Clip Score : 0.30
AI Evaluation
Likelihood of AI : 0.40
Image errors : The image has some noise, particularly in the shadows and the man’s skin. The edges of the image seem blurry.
Lost in the City’s Hustle: A Moment of Anxiety
A woman navigates the bustling city streets, her concerned expression and the blurred background hinting at a sense of unease and isolation. The scene captures a moment of suspense and anxiety amidst the urban chaos.
Prompt
facial-expressions Skepticism: Paranoid, distrustful ; A woman, walking through a crowded street, looking around with suspicion; eye-level; Single Person; Busy city street, people rushing by, street vendors; cinematic
Characteristic
Shot : A woman is looking over her shoulder in a crowded city street, with a blurred background of shops and people. The woman appears to be distressed.
Aesthetic Score : 0.6
Mood : suspenseful, anxious, concerned
Quality
Entropy : 6.87
Noise : 65
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.10
Image errors : There are no visible errors in the image.
Lost in the City’s Shadows
A solitary figure, shrouded in mystery, stands against a backdrop of a bustling cityscape. The warm light casts long shadows, adding to the suspenseful atmosphere. His intense gaze and the blurred background create a sense of intrigue, leaving you wondering what secrets lie hidden within the city’s depths.
Prompt
facial-expressions Skepticism: Isolated, disillusioned ; A hero, standing on a rooftop, looking out at a city skyline, but with a sense of loneliness; eye-level; Hero; City lights, distant sounds of the city; cinematic
Characteristic
Shot : A man in a suit standing in front of a blurry cityscape at dusk.
Aesthetic Score : 0.8
Mood : mysterious, brooding, intense
Quality
Entropy : 6.78
Noise : 74
Prompt Clip Score : 0.20
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image has some minor artifacts, particularly in the background. The bokeh effect is slightly unnatural.
Conclusion
The results of the analysis 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, which is considered below average. This indicates that the generated image’s camera position deviated significantly from the prompt’s instructions.
- Shot Analysis: The model scored 0.58, which is considered good. This means the generated image’s shot composition was fairly close to what was expected based on the prompt.
- Aesthetic Analysis: The model scored 0.06, which is considered average. This suggests that the generated image’s aesthetic was somewhat different from what was expected, but not drastically so.
Overall, the model seems to be better at understanding the scene and shot composition than it is at accurately capturing the desired camera position and aesthetic.
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://deepmind.google/technologies/imagen-2/