Facial expressions are a powerful tool for conveying emotions and intentions in visual storytelling. Generative AI models are increasingly being used to create images with specific facial expressions, but how well do they capture the nuances of human emotion? This blog post delves into the performance of a generative AI model in understanding and generating facial expressions, analyzing its ability to accurately represent camera position, scene details, and aesthetic elements. We’ll explore examples of where the model excels and where it falls short, providing insights into the current state of AI’s ability to create expressive and engaging visuals.
Created with: leonardo-ai
Lost in the City: A Moment of Intensity
A young man stands alone in a bustling city street, his gaze fixed directly on the viewer. The shallow depth of field isolates him, creating a sense of mystery and intrigue. His serious expression suggests a story waiting to be told, a moment of intense emotion captured in this urban landscape.
Prompt
facial-expressions Interest: Intrigued, observant ; A lone figure; eye-level; Single Person; bustling city street; cinematic
Characteristic
Shot : A man with a serious expression is standing on a city street, looking directly at the camera.
Aesthetic Score : 0.7
Mood : serious, moody, urban
Quality
Entropy : 6.67
Noise : 95
Prompt Clip Score : 0.18
AI Evaluation
Likelihood of AI : 0.20
Image errors : No visible artifacts or errors
Superman Faces the Flames: Hero Takes to the Rooftops
A dramatic scene unfolds as a man in a Superman costume stands on a rooftop, overlooking a city with a burning building in the background. The fire and smoke create a sense of urgency and danger, while the Superman’s serious expression suggests heroism and determination.
Prompt
facial-expressions Interest: Focused, determined ; A superhero in a dramatic pose; medium shot; Hero; cityscape with a burning building in the background; cinematic
Characteristic
Shot : A man dressed as Superman stands on a rooftop looking out at a burning building in the distance.
Aesthetic Score : 0.6
Mood : dramatic, serious, heroic
Quality
Entropy : 6.81
Noise : 96
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.20
Image errors : There are some artifacts and errors in the image, such as the subject’s hair. The building appears a bit out of focus, and the edges of the smoke plumes are jagged.
Lost in Thought: A Moment of Tranquility at the Cafe
A woman finds solace in a quiet moment at a cafe, her gaze lost in the window as she holds a book in her lap. The scene evokes a sense of calm and contemplation, inviting you to ponder her thoughts and feelings.
Prompt
facial-expressions Interest: Engrossed, absorbed ; A woman reading a book in a coffee shop; eye-level; Normal People; warm, inviting cafe interior; cinematic
Characteristic
Shot : A woman is sitting in a coffee shop, looking out the window, with a book and a cup of coffee in front of her.
Aesthetic Score : 0.8
Mood : calm, thoughtful, cozy
Quality
Entropy : 6.66
Noise : 95
Prompt Clip Score : 0.30
AI Evaluation
Likelihood of AI : 0.10
Image errors : No visible errors.
Lost in the Game: A Moment of Intense Focus
A young man, bathed in the soft glow of his computer screen, is completely absorbed in a fast-paced video game. The low lighting and his concentrated expression create a palpable sense of suspense and anticipation, drawing the viewer into his world of digital immersion.
Prompt
facial-expressions Interest: Excited, concentrated ; A gamer intensely focused on a screen; close-up; Gamer; dimly lit room with glowing monitor; cinematic
Characteristic
Shot : A young man is sitting at a desk, wearing headphones and looking intently at a computer screen. He appears to be playing a video game.
Aesthetic Score : 0.7
Mood : focused, intense, concentrated
Quality
Entropy : 5.90
Noise : 89
Prompt Clip Score : 0.25
AI Evaluation
Likelihood of AI : 0.10
Image errors : There are no visible errors or artifacts in the image.
A Man’s Melancholy Gaze Through the Rain
A solitary figure sits by a window, his serious expression reflecting a sense of inner turmoil. Rain streaks down the glass, mirroring the melancholic mood as he gazes out, lost in contemplation. The contrast between his outward focus and his internal struggle creates a poignant image of longing and introspection.
Prompt
facial-expressions Interest: Contemplative, thoughtful ; A man gazing out a window at a stormy sky; eye-level; Single Person; dark, moody interior; cinematic
Characteristic
Shot : A man sits by a window, looking out at a rainy day, possibly a field beyond. The lighting is moody.
Aesthetic Score : 0.8
Mood : melancholy, contemplative, introspective
Quality
Entropy : 6.49
Noise : 92
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.10
Image errors : No major image errors, although the window glass may have slight blemishes from the rain drops
Silhouettes of Solitude: A Man Contemplates the City at Sunset
A solitary figure, cloaked in shadow, stands on a rooftop overlooking a sprawling cityscape bathed in the golden hues of sunset. The dramatic interplay of light and shadow highlights the man’s contemplative mood, capturing a moment of urban melancholy.
Prompt
facial-expressions Interest: Confident, determined ; A hero standing on a rooftop overlooking a city; wide shot; Hero; panoramic cityscape with dramatic lighting; cinematic
Characteristic
Shot : A man in a black jacket stands on a rooftop, looking out over a cityscape at sunset.
Aesthetic Score : 0.7
Mood : melancholy, pensive, urban
Quality
Entropy : 6.87
Noise : 90
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.10
Image errors : No visible errors.
Laughter and Warmth: Friends Share a Cozy Meal
Three friends gather around a rustic table, their laughter filling the air. The warm candlelight and inviting setting create a sense of intimacy and comfort, capturing the joy of shared moments with loved ones.
Prompt
facial-expressions Interest: Happy, engaged ; A group of friends laughing together at a dinner table; eye-level; Normal People; cozy, homey dining room; cinematic
Characteristic
Shot : Three friends are sitting at a table in a rustic setting, laughing and enjoying a meal together.
Aesthetic Score : 0.7
Mood : happy, relaxed, warm
Quality
Entropy : 6.68
Noise : 100
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.10
Image errors : There are some minor artifacts in the image, such as noise and banding. The colors are also a bit muted, and the image could benefit from more contrast.
Lost in the Code: A Moment of Intense Focus
A young man, headphones on, is completely absorbed in his work. The dimly lit room and dramatic composition emphasize his concentration, hinting at the importance of his task. This image captures the essence of dedication and the power of focus.
Prompt
facial-expressions Interest: Thrilled, focused ; A gamer’s hands rapidly moving across a keyboard and mouse; close-up; Gamer; brightly lit gaming setup with flashing lights; cinematic
Characteristic
Shot : A young man wearing headphones is sitting in a dimly lit room, focused on a computer screen. He appears to be playing a game or working on a project.
Aesthetic Score : 0.7
Mood : focused, intense, serious
Quality
Entropy : 6.30
Noise : 91
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.20
Image errors : No visible errors
Lost in Art: A Moment of Contemplation
A woman stands captivated by a painting in a dimly lit gallery, her focus creating a sense of intrigue. The darkened walls and rich wood floor enhance the artistic atmosphere, inviting viewers to share in her contemplative moment.
Prompt
facial-expressions Interest: Appreciative, curious ; A woman looking at a painting in a museum; eye-level; Single Person; grand museum hall with intricate artwork; cinematic
Characteristic
Shot : A woman stands in a museum gallery, looking at a painting, in front of several other paintings.
Aesthetic Score : 0.6
Mood : calm, contemplative, thoughtful
Quality
Entropy : 6.59
Noise : 98
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.10
Image errors : No notable artifacts, but slight sharpness problems. The lighting is a bit flat.
Gritty Urban Warfare: Soldier’s Muzzle Flash Illuminates the Danger
A soldier, clad in tactical gear, unleashes a torrent of firepower in a dimly lit urban environment. The muzzle flash and billowing smoke create a dramatic and intense scene, highlighting the raw power and danger of the moment. The soldier’s focused expression adds to the intensity, capturing the heart-pounding action of the situation.
Prompt
facial-expressions Interest: Intense, focused ; A hero facing off against a villain; medium shot; Hero; dramatic, action-packed scene with explosions and smoke; cinematic
Characteristic
Shot : A man in military gear is firing a weapon in a smoky environment. He looks intense and focused. There is a background out of focus.
Aesthetic Score : 0.7
Mood : intense, dramatic, action
Quality
Entropy : 6.59
Noise : 91
Prompt Clip Score : 0.23
AI Evaluation
Likelihood of AI : 0.30
Image errors : The image appears to have been sharpened too aggressively, leading to a slightly artificial appearance. The lighting also appears a bit uneven.
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 0.16, which is below the “good” range of 0.5 to 0.75. This indicates that the model didn’t accurately capture the intended camera position in the prompt.
- Shot Analysis: The model scored 0.49, which is also below the “good” range. This suggests 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 0.12, which is within the “very good” range of -0.2 to 0.1. This means the generated image’s aesthetic closely matched the expected aesthetic described in the prompt.
Overall: While the model excelled in capturing the desired aesthetic, it struggled with accurately representing the camera position and scene described in the prompt.