AI's Facial Expressions: A Mixed Bag of Success with Titan-g1
- 10 minutes read - 1950 wordsTable of Contents
Facial expressions are a powerful tool in storytelling, conveying a wide range of emotions and adding depth to characters. In the realm of AI image generation, capturing these expressions accurately is crucial for creating compelling and realistic visuals. This blog post explores the results of an experiment where an AI model was tasked with generating images based on detailed scene descriptions, focusing on the model’s ability to understand and depict facial expressions. We’ll delve into the specific areas where the model excelled and where it needs improvement, providing insights into the current capabilities and limitations of AI in image generation.
Created with: titan-g1
Lost in the Shadows: A Woman’s Solitary Moment in a Gloomy Alley
A woman sits alone in a narrow, graffiti-covered alleyway, her gaze fixed on the distant light. The dim lighting and cluttered surroundings create a sense of loneliness and mystery, leaving the viewer to wonder about her thoughts and the story behind her solitude.
Prompt
facial-expressions Disgust: Despair and alienation ; A lone figure, hunched over in a dimly lit alleyway; eye-level; Single Person; overflowing trash bins and graffiti-covered walls; cinematic
Characteristic
Shot : A woman in a dark jacket sits in a narrow alleyway with overflowing garbage cans. The walls are gray and covered in graffiti. It is a gloomy and somewhat depressing scene, evoking feelings of neglect and isolation.
Aesthetic Score : 0.4
Mood : gloomy, isolated, neglected
Quality
Entropy : 6.93
Noise : 106
Prompt Clip Score : 0.25
AI Evaluation
Likelihood of AI : 0.20
Image errors : Slight blurring around the woman’s edges, slight moire pattern on the wall
A Moment of Artistic Revelation: Man’s Shock at Abstract Masterpiece
A man stands transfixed before a vibrant, abstract painting, his expression a mixture of surprise and intrigue. The blurry, chaotic brushstrokes of the artwork seem to mirror the man’s internal turmoil, creating a sense of dramatic tension. Is he captivated by the beauty, or overwhelmed by the complexity? This moment of artistic revelation leaves us questioning the power of art to evoke such strong emotions.
Prompt
facial-expressions Disgust: Horror and disgust ; A seasoned detective, his face etched with disgust, stares at a vandalized masterpiece in a museum; eye-level; Detective; a chaotic scene with paint splattered across the canvas and broken glass on the floor.; cinematic
Characteristic
Shot : A man is standing in front of a painting, looking at it in surprise.
Aesthetic Score : 0.6
Mood : shocked, tense, curious
Quality
Entropy : 6.92
Noise : 104
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.10
Image errors : No noticeable artifacts or errors.
A Moment of Kitchen Despair
A woman stands in her kitchen, her face etched with sadness and frustration. A large pot and a plate of food sit before her, but the scene lacks visual excitement. The woman’s emotional state is the primary focus, creating a sense of quiet, internal struggle.
Prompt
facial-expressions Disgust: Disappointment and disgust ; A young woman, her face pale and wrinkled, as she stares at a plate of spoiled food; eye-level; Normal Person; a cluttered kitchen with dirty dishes and a overflowing trash can; cinematic
Characteristic
Shot : A woman is standing in a kitchen looking distressed, with a plate of food and a messy sink behind her.
Aesthetic Score : 0.4
Mood : distressed, overwhelmed, frustration
Quality
Entropy : 6.93
Noise : 100
Prompt Clip Score : 0.30
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is slightly blurred and the lighting is uneven. There is also a slight glare on the woman’s hair.
Awe-Inspiring Waterfall: A Moment of Wonder in Nature
A man, captivated by the sheer power of a cascading waterfall, stands in awe, his backpack and blue jacket a testament to his adventurous spirit. The scene evokes a sense of wonder and excitement, highlighting the majesty of nature.
Prompt
facial-expressions Disgust: Unease ; A hiker, their eyes wide with disbelief, as they stumble upon a breathtaking vista in the middle of a dense forest; eye-level; Hiker; a sun-drenched clearing with towering trees and a cascading waterfall; cinematic
Characteristic
Shot : A man in a green jacket with a backpack is standing in front of a waterfall, looking up in awe. The surrounding scenery is a mix of greenery and rocks.
Aesthetic Score : 0.6
Mood : awe, wonder, adventure
Quality
Entropy : 6.68
Noise : 106
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.20
Image errors : No major errors, but the lighting could be improved, particularly on the man’s face.
A Walk into the Unknown
A woman, shrouded in mystery, walks away from a pile of discarded tires. The low angle and blurry background create a sense of unease, leaving the viewer wondering what secrets lie ahead. The image evokes a dark and tense mood, hinting at a story waiting to unfold.
Prompt
facial-expressions Disgust: Repulsion ; A woman, her face contorted in disgust, as she walks past a pile of discarded tires; eye-level; Single Person; a dusty and neglected parking lot with overflowing recycling bins.; cinematic
Characteristic
Shot : A woman walks past a large pile of old tires in a parking lot, her expression is concerned, possibly due to the unsettling nature of the environment.
Aesthetic Score : 0.4
Mood : tense, moody, unsettling
Quality
Entropy : 6.80
Noise : 99
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image appears to be properly exposed, with good sharpness and color balance. There are no visible technical errors.
Street Confrontation: Two Men Locked in Heated Argument
A tense moment unfolds on a city street as two men engage in a heated argument. The man in the brown coat points accusingly, while the other man appears surprised and defensive. The image captures the raw emotion and intensity of the confrontation, leaving the viewer wondering what sparked the dispute.
Prompt
facial-expressions Disgust: disgust ; A seasoned detective, his face etched, as he confronts a con artist who has pulled off a daring heist; eye-level; Detective; a bustling marketplace with a towering, flamboyant figure in a brightly colored suit.; cinematic
Characteristic
Shot : Two men in suits are engaged in a tense confrontation in an urban setting.
Aesthetic Score : 0.6
Mood : dramatic, confrontational, serious
Quality
Entropy : 6.89
Noise : 101
Prompt Clip Score : 0.23
AI Evaluation
Likelihood of AI : 0.20
Image errors : There are no noticeable artifacts or errors in the image.
Disgust and Fear in the Kitchen: A Missed Opportunity for Dramatic Impact
This image captures a woman’s visceral reaction to a rat in the kitchen, conveying feelings of disgust, surprise, and fear. However, the framing and composition fail to fully capitalize on the dramatic potential of the scene, leaving the viewer with a sense of missed opportunity.
Prompt
facial-expressions Disgust: Horror and disgust ; A family, their faces twisted in disgust, as they discover a dead rat in their kitchen; eye-level; Normal People; a cluttered and messy kitchen with dirty dishes and a overflowing trash can; cinematic
Characteristic
Shot : A woman is reacting in horror to a rat in a kitchen. A man stands in the background, seemingly unaware of the situation.
Aesthetic Score : 0.3
Mood : horror, shock, disgust
Quality
Entropy : 6.92
Noise : 103
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is slightly overexposed and has some graininess. The background is also a bit cluttered.
The Thrill of Victory: Gamer’s Excitement Captured in a Moment
This image captures the raw excitement of a gamer as he plays, his expression and gestures conveying the intensity of the moment. While the framing is static, the energy of the scene is undeniable, with a hint of playful camaraderie from the partially obscured figure in the background.
Prompt
facial-expressions Disgust: Fear and disgust ; A gamer, their face pale and sweaty, as they witness a disturbing scene in a horror game; eye-level; Gamer; a dimly lit gaming room with a flickering monitor and a dark, ominous atmosphere; cinematic
Characteristic
Shot : A young man is playing a video game. He is wearing headphones and is very excited. He is looking at the screen in front of him. There is another man in the background who is also excited.
Aesthetic Score : 0.6
Mood : intense, focused, excited
Quality
Entropy : 6.72
Noise : 101
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is slightly grainy, especially in the darker areas.
Picnic Surprise: Woman Startled by Spider in Salad
A woman enjoying a peaceful picnic is suddenly met with a horrifying surprise - a spider lurking in her salad. The image captures her startled reaction, with the spider framed against her face, creating a sense of impending danger and disgust.
Prompt
facial-expressions Disgust: Revulsion and disgust ; A woman, her face contorted, as she discovers a spider in her salad; eye-level; Single Person; a brightly lit picnic table in a park with a basket full of food and a spider crawling on a plate; cinematic
Characteristic
Shot : A woman is startled by a spider in her salad during a picnic.
Aesthetic Score : 0.3
Mood : startled, disgusted, humorous
Quality
Entropy : 6.58
Noise : 97
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image appears to be slightly overexposed, especially in the background.
The Focus of the Game
A man in a plaid shirt, bathed in the soft glow of a green lamp, meticulously lines up his shot on the pool table. The dimly lit room adds to the sense of anticipation as he prepares to sink the white ball, his concentration unwavering.
Prompt
facial-expressions Disgust: disgust ; A seasoned pool player, eyes narrowed with suspicion, confronts a young up-and-comer in a smoky, dimly lit pool hall; eye-level; Pool Shark; cinematic
Characteristic
Shot : A man is playing pool in a dimly lit pub. The man is focused on the shot, and the balls are arranged on the green felt table. A blurry figure watches from the background.
Aesthetic Score : 0.6
Mood : focused, intense, casual
Quality
Entropy : 6.72
Noise : 108
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.10
Image errors : There are no obvious errors in the image. The lighting is a bit dark and uneven, but this could be intentional.
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 0.23, which is below the “good” range of 0.5 to 0.75. This suggests the model didn’t fully capture the intended camera position in the prompt.
- Shot Analysis: The model scored 0.68, which falls within the “good” range. This indicates the model was able to understand the scene described in the prompt fairly well.
- Aesthetic Analysis: The model scored 0.31, which is significantly higher than the “very good” range of -0.2 to 0.1. This suggests the generated image’s aesthetic deviated significantly from the expected aesthetic described in the prompt.
Overall, the model shows promise in understanding scene composition and camera angles, but needs improvement in generating images that match the desired 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://docs.aws.amazon.com/bedrock/latest/userguide/titan-image-models.html