AI Captures the Emotion, But Misses the Angle: A Look at Facial Expressions in AI-Generated Images with Imagen-v2
- 9 minutes read - 1753 wordsTable of Contents
Facial expressions are a powerful tool for conveying emotions and intentions in visual storytelling. Dramatic facial expressions, in particular, can heighten the impact of a scene and draw the viewer in. Generative AI models are increasingly being used to create realistic facial expressions in images and videos, but how well do they capture the nuances of human emotion and the complexities of camera positioning? This blog post explores the capabilities of a generative AI model in creating images with dramatic facial expressions, analyzing its performance across a range of scenes and camera angles.
Created with: imagen-v2
Lost in Thought: A Moment of Melancholy by the Window
A woman sits by a window, her gaze lost in the blurred snowy scene outside. The soft lighting and her thoughtful expression evoke a sense of melancholy and introspection, leaving the viewer to ponder her thoughts and the mystery surrounding her.
Prompt
facial-expressions Worry: melancholy, lonely ; Single woman; eye-level; Single Persons; dimly lit coffee shop with rain outside; cinematic
Characteristic
Shot : A woman sits by a window, looking out, with a thoughtful expression. The background is out of focus and shows a snowy scene through a window.
Aesthetic Score : 0.7
Mood : melancholy, contemplative, introspective
Quality
Entropy : 6.58
Noise : 100
Prompt Clip Score : 0.25
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image is slightly blurry, especially in the background. There are some artifacts in the woman’s hair, and the color balance is a bit off. There is a slight color cast, that gives the image a greenish tint
The Hero’s Gaze: A Moment of Intensity
A close-up portrait of a superhero, their blue and red costume stark against the blurry city lights. Their intense gaze draws you in, promising a story of drama, mystery, and suspense.
Prompt
facial-expressions Worry: intense, burdened ; Man in a superhero costume; medium shot; Heroes; cityscape at night with flashing sirens; cinematic
Characteristic
Shot : A superhero in a blue and red costume, wearing a mask, with a city lights background.
Aesthetic Score : 0.6
Mood : intense, serious, hopeful
Quality
Entropy : 6.67
Noise : 111
Prompt Clip Score : 0.22
AI Evaluation
Likelihood of AI : 0.70
Image errors : The edges of the mask and the subject’s hair seem slightly blurry and unnatural, potentially due to post-processing.
Lost in the Crowd: A Moment of Anxiety
A young woman, her face etched with worry, gazes upwards in a crowded, bustling environment. The blur of the background emphasizes her isolation and the weight of her anxieties. Her mustard yellow scarf and blue shirt offer a splash of color against the muted tones of her surroundings, highlighting the contrast between her inner turmoil and the vibrant world around her.
Prompt
facial-expressions Worry: anxious, overwhelmed ; Young woman in a crowded subway; eye-level; Normal People; blurred faces of commuters; cinematic
Characteristic
Shot : A young woman with long brown hair is standing in a crowded subway car, looking up with a worried expression.
Aesthetic Score : 0.6
Mood : anxious, tense, uncertain
Quality
Entropy : 6.60
Noise : 60
Prompt Clip Score : 0.31
AI Evaluation
Likelihood of AI : 0.80
Image errors : The hair texture and details in the background appear slightly blurry, possible rendering errors.
Lost in the Rhythm: A Portrait of Intensity
A close-up shot captures a man immersed in his music, his face bathed in vibrant red and blue light. His focused expression speaks volumes about the power of the moment, creating a dramatic and visually captivating scene.
Prompt
facial-expressions Worry: intense, focused ; Gamer with headphones on; close-up; Gamer; dimly lit room with glowing computer screen; cinematic
Characteristic
Shot : A close-up shot of a man wearing headphones, with red and blue lighting illuminating his face, he is looking directly at the camera with an intense expression.
Aesthetic Score : 0.7
Mood : intense, dramatic, mysterious
Quality
Entropy : 6.10
Noise : 68
Prompt Clip Score : 0.32
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image has some artifacts around the edges, particularly in the hair. There is also some blurring in the background.
Autumn Contemplation: A Moment of Solitude in the Park
A man sits alone on a bench, lost in thought amidst a sea of fallen leaves. The soft lighting and muted colors of autumn create a melancholic atmosphere, evoking a sense of loneliness and contemplation. The blurred background and the solitary figure in the foreground emphasize the man’s isolation and introspective mood.
Prompt
facial-expressions Worry: sad, reflective ; Man sitting alone on a park bench; long shot; Single Persons; empty park with falling leaves; cinematic
Characteristic
Shot : A man is sitting on a bench in a park, looking thoughtful and sad. The background is a blurry image of a tree with yellow leaves.
Aesthetic Score : 0.6
Mood : melancholy, contemplative, sad
Quality
Entropy : 6.87
Noise : 91
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.30
Image errors : No noticeable artifacts, the image is slightly overexposed, the lighting is flat and not very dynamic
A City on the Brink: One Woman’s Determined Gaze
A young woman stands resolute, her worried expression fixed on the camera. Behind her, a city shrouded in smoke hints at a looming threat. The dramatic scene evokes a sense of tension and melancholic anticipation, leaving the viewer questioning what lies ahead.
Prompt
facial-expressions Worry: determined, resolute ; Heroine standing on a rooftop; medium shot; Heroes; cityscape with smoke and fire in the distance; cinematic
Characteristic
Shot : A woman in a red leather jacket stands in front of a cityscape with smoke in the background.
Aesthetic Score : 0.7
Mood : dramatic, suspenseful, intense
Quality
Entropy : 6.38
Noise : 76
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.20
Image errors : The smoke in the background looks slightly artificial and the color grading is a bit too saturated.
A Tense Moment in the Kitchen
A young couple faces a difficult conversation, their expressions revealing a mix of anger, disappointment, and sadness. The messy table and dramatic lighting amplify the tension in the air, hinting at a troubled relationship.
Prompt
facial-expressions Worry: tense, frustrated ; Couple arguing in a kitchen; eye-level; Normal People; cluttered kitchen with dirty dishes; cinematic
Characteristic
Shot : A couple sits at a kitchen table, the woman is standing and looking at the man, who is sitting. There are dirty dishes on the table and a window behind the man.
Aesthetic Score : 0.6
Mood : tense, awkward, serious
Quality
Entropy : 6.71
Noise : 84
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is slightly grainy. There is a slight blur in the background. The lighting is uneven. The composition is slightly off-center.
Lost in the Game: A Moment of Intense Focus
A young man, headphones on, is completely immersed in a video game. The blurred background of red and blue lights adds to the sense of action and excitement, highlighting his intense focus and determination.
Prompt
facial-expressions Worry: intense, focused ; Gamer’s hands on a keyboard; close-up; Gamer; flashing lights and sounds from the game; cinematic
Characteristic
Shot : A young man wearing headphones is intensely focused on something out of frame, possibly a video game, with dramatic lighting and blurred background.
Aesthetic Score : 0.6
Mood : intense, focused, dramatic
Quality
Entropy : 6.42
Noise : 76
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.80
Image errors : The background appears overly blurred and unrealistic. The skin texture looks a bit artificial, especially around the eyes and lips. The lighting is harsh and creates a lot of contrast.
Lost in the City Lights: A Moment of Melancholy
A woman stands alone in the night, her gaze lost in the distance. The soft lighting and shallow depth of field create a sense of intimacy and vulnerability, capturing a moment of quiet sadness.
Prompt
facial-expressions Worry: lonely, vulnerable ; Woman walking alone at night; long shot; Single Persons; deserted street with streetlights; cinematic
Characteristic
Shot : A woman, looking troubled, is standing in the street with blurred out lights in the background.
Aesthetic Score : 0.6
Mood : melancholy, pensive, sad
Quality
Entropy : 6.38
Noise : 105
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.20
Image errors : No visible errors.
The Face of War: A Soldier’s Gritty Determination
A close-up portrait captures the intensity of a futuristic soldier, his face etched with dirt and determination. The blurred, war-torn backdrop adds to the sense of urgency and danger, creating a powerful and somber image.
Prompt
facial-expressions Worry: serious, strategic ; Hero looking at a map; medium shot; Heroes; war-torn battlefield with smoke and debris; cinematic
Characteristic
Shot : A close-up portrait of a soldier with a dirty face, looking serious and determined, with a blurry background of a war-torn landscape. The scene evokes a sense of conflict, danger, and grit.
Aesthetic Score : 0.7
Mood : serious, determined, gritty
Quality
Entropy : 6.80
Noise : 69
Prompt Clip Score : 0.25
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image exhibits a slight blurriness around the edges, particularly noticeable in the background, which may suggest some image processing or editing artifacts.
Conclusion
The results of the analysis show that the generative AI model performed well in terms of understanding the scene and camera position, but struggled with the aesthetic aspect. Here’s a breakdown:
- Camera Position: The model scored 0.4, which is considered below average. This suggests that the model didn’t accurately capture the intended camera position described in the prompt.
- Shot Analysis: The model scored 0.595, which is considered good. This indicates that the model was able to understand the scene described in the prompt and create a shot that aligns with it.
- Aesthetic Analysis: The model scored 0.1, which is considered very good. This means that the generated image closely matched the expected aesthetic style.
Overall, the model demonstrates a good understanding of the scene and its aesthetic, but needs improvement in accurately capturing the intended camera position.
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/