AI's Facial Expressions: A Mixed Bag of Success with Stable-diffusion
- 9 minutes read - 1771 wordsTable of Contents
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 across a range of scenes, exploring its strengths and weaknesses in capturing camera position, shot composition, and aesthetic expectations. We’ll examine examples where the model excels and where it falls short, providing insights into the current state of AI-generated facial expressions.
Created with: stability-ai-core
Lost in Thought: A Moment of Quiet Contemplation
A young woman finds solace in a cozy cafe, her thoughtful gaze fixed on the world outside. The soft lighting and her serene expression create a sense of intimacy and quiet contemplation, capturing the essence of a peaceful moment.
Prompt
facial-expressions Gratitude: Contentment and appreciation for solitude ; Single woman; eye-level; Single Persons; cozy cafe with warm lighting; cinematic
Characteristic
Shot : A woman in a cafe, sitting at a table with a cup of coffee. She is looking off-camera with a thoughtful expression. The cafe is warm and inviting, with soft lighting and a relaxed atmosphere.
Aesthetic Score : 0.7
Mood : calm, contemplative, cozy
Quality
Entropy : 6.68
Noise : 76
Prompt Clip Score : 0.22
AI Evaluation
Likelihood of AI : 0.10
Image errors : No significant errors or artifacts
Heroic Firefighter Rescues Child from Burning Building
A dramatic scene unfolds as a firefighter in full gear carries a young child to safety from a raging inferno. The contrast between the flames and the firefighter’s protective presence creates a powerful image of heroism and resilience.
Prompt
facial-expressions Gratitude: Relief, gratitude for the hero’s bravery ; Firefighter rescuing a child from a burning building; wide shot; Heroes; smoke and flames in the background; cinematic
Characteristic
Shot : A fireman carrying a child away from a burning building. The fire is in the background and the fireman is in the foreground.
Aesthetic Score : 0.7
Mood : dramatic, heroic, hopeful
Quality
Entropy : 6.85
Noise : 74
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.10
Image errors : No noticeable artifacts, but the image is slightly blurry, particularly the background fire.
The Joy of a Family Meal
A heartwarming scene of a family gathered around a table, sharing a meal and laughter. The warm lighting and genuine smiles create a sense of intimacy and happiness, capturing the essence of family togetherness.
Prompt
facial-expressions Gratitude: Warmth, appreciation for family and connection ; Family having dinner together; eye-level; Normal People; warm, inviting kitchen; cinematic
Characteristic
Shot : A family is enjoying a meal together at a dinner table in a kitchen. It is a warm and inviting scene, with candles and soft lighting.
Aesthetic Score : 0.7
Mood : warm, happy, togetherness
Quality
Entropy : 6.74
Noise : 77
Prompt Clip Score : 0.21
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is slightly blurry, especially in the background.
The Joy of Victory: Gamer’s Enthusiasm Captures the Screen
A young man, radiating pure joy, is fully immersed in a game, his smile and focused expression captured in a close-up shot. The dimly lit room, filled with computer monitors, adds to the intensity of the moment, highlighting the gamer’s passion and energy.
Prompt
facial-expressions Gratitude: Excitement, gratitude for the shared experience ; Gamer celebrating a victory with friends; close-up; Gamer; brightly lit gaming room with screens and controllers; cinematic
Characteristic
Shot : A young man wearing headphones is smiling and looking directly at the camera, while sitting in a gaming chair and typing on a keyboard. Two other people are sitting in the background, out of focus.
Aesthetic Score : 0.7
Mood : happy, focused, energetic
Quality
Entropy : 6.36
Noise : 65
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.10
Image errors : There are a few minor image errors, including some noise in the shadows and a slight halo effect around the subject’s head.
Silhouetted in Golden Light: A Moment of Contemplation
A man, lost in thought, stands amidst a field of tall grass as the sun dips below the horizon. The soft, golden light casts a melancholic yet peaceful mood, inviting viewers to share in his quiet contemplation.
Prompt
facial-expressions Gratitude: Awe, gratitude for the beauty of nature ; Man looking out at a beautiful sunset; eye-level; Single Persons; vast, open field with golden light; cinematic
Characteristic
Shot : A man is standing in a field of corn, looking off into the distance at a sunset. He is wearing a denim jacket and a t-shirt.
Aesthetic Score : 0.7
Mood : melancholy, contemplative, nostalgic
Quality
Entropy : 6.79
Noise : 71
Prompt Clip Score : 0.20
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image appears to be slightly overexposed, and there is some noise in the shadows.
Serious Discussion in the Hospital Room
A tense atmosphere fills the hospital room as a doctor speaks with a patient. The serious expressions on their faces and the presence of another doctor in the background create a sense of suspense and drama. The scene evokes a feeling of concern and anticipation, leaving the viewer wondering what the outcome will be.
Prompt
facial-expressions Gratitude: Hope, gratitude for the doctor’s care ; Doctor comforting a patient; medium shot; Heroes; sterile hospital room with medical equipment; cinematic
Characteristic
Shot : A hospital room with a doctor talking to a patient. There is a nurse in the background.
Aesthetic Score : 0.6
Mood : serious, concerned, tense
Quality
Entropy : 6.89
Noise : 63
Prompt Clip Score : 0.21
AI Evaluation
Likelihood of AI : 0.20
Image errors : There are some minor artifacts in the image, particularly around the edges of the objects. The lighting is a bit uneven, which casts shadows on some of the faces.
Sunny Days and Happy Friends: A Perfect Picnic
Capture the joy of a sunny day with this heartwarming image of friends enjoying a picnic in the park. The vibrant colors and cheerful expressions create a sense of warmth and lightheartedness, perfect for evoking feelings of friendship and happiness.
Prompt
facial-expressions Gratitude: Joy, gratitude for friendship and good times ; Group of friends laughing together at a picnic; eye-level; Normal People; sunny park with green grass and trees; cinematic
Characteristic
Shot : Four friends laughing and enjoying a picnic in a park on a sunny day.
Aesthetic Score : 0.8
Mood : joyful, carefree, relaxed
Quality
Entropy : 6.84
Noise : 85
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.20
Image errors : None
Sharing the Joy: A Moment of Happiness Captured
This image captures a man’s genuine smile as he holds a device, radiating a sense of happiness and connection. The blurred background suggests a lively atmosphere, adding to the overall feeling of joy and camaraderie.
Prompt
facial-expressions Gratitude: Pride, gratitude for recognition and hard work ; Gamer receiving an award for their achievements; close-up; Gamer; stage with a crowd and flashing lights; cinematic
Characteristic
Shot : A young man is holding a small object in his hand and smiling at the camera. He is wearing a black shirt with a rainbow lanyard and an Adidas logo. He is surrounded by other people who are blurred out of focus. The background is a stage with colorful lights and screens.
Aesthetic Score : 0.7
Mood : happy, friendly, confident
Quality
Entropy : 6.46
Noise : 67
Prompt Clip Score : 0.18
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image has a slight amount of noise and some blurriness in the background.
Finding Peace in the Pages: A Moment of Tranquility in the Library
A woman finds solace and focus amidst towering bookshelves, bathed in warm light. Her contemplative expression and the serene atmosphere evoke a sense of calm and quiet reflection.
Prompt
facial-expressions Gratitude: Peace, gratitude for knowledge and escape ; Woman reading a book in a quiet library; eye-level; Single Persons; peaceful library with bookshelves and natural light; cinematic
Characteristic
Shot : A woman is sitting in a library, reading a book, with a bookshelf behind her.
Aesthetic Score : 0.7
Mood : calm, cozy, introspective
Quality
Entropy : 6.55
Noise : 65
Prompt Clip Score : 0.20
AI Evaluation
Likelihood of AI : 0.10
Image errors : No noticeable errors.
One Man’s Trash, Another’s Treasure: Beach Cleanup Inspires Hope
A man, radiating positivity, kneels on a sandy beach, diligently picking up trash with a blue bucket and yellow shovel. His genuine smile and focused effort, amidst a backdrop of other volunteers, create a heartwarming scene of community action and environmental stewardship. This image embodies hope and the power of collective action to make a difference.
Prompt
facial-expressions Gratitude: Satisfaction, gratitude for making a difference ; Volunteer helping to clean up a beach; wide shot; Heroes; beautiful beach with clear water and blue sky; cinematic
Characteristic
Shot : A man is kneeling on the beach, collecting garbage in a blue bucket. Other people are in the background, also cleaning the beach.
Aesthetic Score : 0.7
Mood : happy, positive, community
Quality
Entropy : 6.78
Noise : 67
Prompt Clip Score : 0.23
AI Evaluation
Likelihood of AI : 0.20
Image errors : No visible artifacts or errors.
Conclusion
The results show that the generative AI model performed well in understanding the camera position and shot composition, but struggled with the aesthetic expectations. Here’s a breakdown:
- Camera Position: The model scored 0.38, which is below the “good” range of 0.5 to 0.75. This suggests that the model didn’t fully capture the intended camera position described in the prompt.
- Shot Analysis: The model scored 0.535, which falls within the “good” range. This indicates that the model was able to understand and translate the scene description in the prompt into a visually coherent shot.
- Aesthetic Analysis: The model scored 0.09, which is significantly higher than the “very good” range of -0.2 to 0.1. This suggests that the generated image’s aesthetic deviated significantly from the expected aesthetic described in the prompt.
Overall, the model demonstrated a good understanding of shot composition but struggled with camera positioning and aesthetic expectations.
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://stability.ai