AI's Facial Expressions: A Mixed Bag of Success with Leonardo-ai
- 9 minutes read - 1784 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 perform? This blog post delves into the performance of a generative AI model in capturing facial expressions across diverse scenes, analyzing its strengths and weaknesses in understanding camera position, scene description, and aesthetic elements. We’ll explore examples where the model excels and where it struggles, providing insights into the current state of AI-generated facial expressions.
Created with: leonardo-ai
Lost in Thought on a City Street
A young woman, shrouded in a leather jacket, stands alone on a city street at dusk. Her gaze is fixed on something beyond the frame, her expression thoughtful and melancholic. The blurred background adds to the sense of isolation and mystery, drawing the viewer into her world of contemplation.
Prompt
facial-expressions Daydreaming: Melancholy, lost in thought ; A lone figure; eye-level; Single Person; bustling city street; cinematic
Characteristic
Shot : A young woman, dressed in a leather jacket, stands on a city street at night, looking off to the side. The city lights are blurred in the background, creating a sense of depth and distance.
Aesthetic Score : 0.7
Mood : melancholy, pensive, urban
Quality
Entropy : 6.60
Noise : 92
Prompt Clip Score : 0.19
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image is slightly blurry, but this could be an intentional stylistic choice. The colors are muted and the overall tone of the image is dark.
Superman Stands Watch Over a City Awash in Lights
A solitary figure against the night sky, Superman surveys the city below, his cape billowing in the wind. The image captures the hero’s power and isolation, a stark reminder of the responsibility he carries to protect the innocent.
Prompt
facial-expressions Daydreaming: Confident, determined ; A superhero standing on a rooftop; high angle; Hero; cityscape at night; cinematic
Characteristic
Shot : A man dressed as a superhero stands on a rooftop overlooking a cityscape at night, his face serious as he gazes out into the distance.
Aesthetic Score : 0.7
Mood : dramatic, mysterious, hopeful
Quality
Entropy : 6.18
Noise : 86
Prompt Clip Score : 0.22
AI Evaluation
Likelihood of AI : 0.30
Image errors : The image has some minor artifacts, particularly around the edges of the subject’s cape and the buildings in the background.
Lost in Thought: A Moment of Quiet Contemplation
A woman finds solace in a moment of quiet reflection, lost in thought as she gazes out the window of a cozy cafe. The warm glow of the setting sun casts a soft light on her face, highlighting the pensive mood she embodies. The image evokes a sense of calm and introspection, inviting viewers to share in her quiet contemplation.
Prompt
facial-expressions Daydreaming: Peaceful, content ; A woman sipping coffee in a cafe; eye-level; Normal People; warm, inviting cafe interior; cinematic
Characteristic
Shot : A woman is sitting in a cafe, looking out the window. She is holding a cup of coffee. The background is blurry. The overall lighting is warm and inviting.
Aesthetic Score : 0.7
Mood : pensive, relaxed, contemplative
Quality
Entropy : 6.79
Noise : 92
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image has some slight noise and grain, particularly in the shadows. There is a slight blur in the background, which is likely due to the camera’s depth of field.
Lost in the Code: A Man’s Intense Focus Under Dimly Lit Shadows
A solitary figure, headphones on, is captivated by the glow of his computer screen. The dimly lit room casts long shadows, adding an air of mystery and intrigue to his focused gaze. This image evokes a sense of intense concentration and contemplation, leaving the viewer wondering what secrets lie within the code.
Prompt
facial-expressions Daydreaming: Engrossed, excited ; A gamer intensely focused on a screen; close-up; Gamer; dimly lit room with gaming peripherals; cinematic
Characteristic
Shot : A young man wearing headphones is sitting in front of a computer screen. He is looking at the screen with a focused expression. The room is dimly lit, and the only light source is the screen.
Aesthetic Score : 0.7
Mood : focused, serious, intense
Quality
Entropy : 6.06
Noise : 89
Prompt Clip Score : 0.22
AI Evaluation
Likelihood of AI : 0.20
Image errors : None
Lost in Thought: A Boy’s Moment of Contemplation
A young boy sits by a window, bathed in soft light, his thoughtful expression hinting at a world of introspection. The scene evokes a sense of quiet contemplation and invites viewers to share in his pensive mood.
Prompt
facial-expressions Daydreaming: Curious, imaginative ; A child staring out a window; eye-level; Single Person; lush green garden; cinematic
Characteristic
Shot : A young boy is sitting by a window, looking up and out, lost in thought.
Aesthetic Score : 0.7
Mood : pensive, contemplative, nostalgic
Quality
Entropy : 6.90
Noise : 96
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.20
Image errors : No noticeable artifacts or errors.
A Knight’s Journey: Sun-Drenched Valor in the Forest
A knight in shining armor rides through a sun-drenched forest, his presence both heroic and mysterious. The warm glow of the sun creates a dramatic effect, highlighting his power and setting the stage for an epic adventure.
Prompt
facial-expressions Daydreaming: Brave, adventurous ; A knight in shining armor riding through a forest; wide shot; Hero; mystical forest with dappled sunlight; cinematic
Characteristic
Shot : A knight in full armor is riding a horse through a forest. The sun is shining through the trees, creating a magical atmosphere.
Aesthetic Score : 0.7
Mood : mysterious, epic, adventurous
Quality
Entropy : 6.72
Noise : 106
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.20
Image errors : There are no visible image errors.
Laughter and Sunshine: Friends Enjoy a Perfect Picnic
Capture the joy of friendship with this heartwarming image of three friends sharing a picnic in a sunny park. The warm colors and genuine laughter create a sense of happiness and lightheartedness, making this a perfect picture for anyone who appreciates the simple pleasures of life.
Prompt
facial-expressions Daydreaming: Joyful, carefree ; A group of friends laughing together at a picnic; eye-level; Normal People; sunny park with picnic blanket; cinematic
Characteristic
Shot : Three friends enjoying a picnic in a park, sitting on a red and white checkered blanket. The sun is shining and they are laughing.
Aesthetic Score : 0.7
Mood : joyful, cheerful, carefree
Quality
Entropy : 6.88
Noise : 102
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image quality is good, there are no visible artifacts or errors.
Lost in the Glow: A Young Man’s Intense Focus Under RGB Lighting
A dimly lit room, a young man with headphones, and a keyboard bathed in vibrant RGB light. This scene captures the intense focus and dedication of a person lost in their work, creating an atmosphere of mystery and intrigue.
Prompt
facial-expressions Daydreaming: Thrilled, competitive ; A gamer’s hands rapidly moving across a keyboard; close-up; Gamer; brightly lit gaming setup with glowing screen; cinematic
Characteristic
Shot : A young man wearing a headset is focused on playing a game on a computer, his hands are on the keyboard, and the screen in the background is blurred. It’s a nighttime scene.
Aesthetic Score : 0.6
Mood : intense, focused, competitive
Quality
Entropy : 6.24
Noise : 93
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.20
Image errors : No visible image errors.
Solitude by the Sea: A Moment of Tranquility
A woman finds peace as she walks along a serene beach, the vast ocean and a distant figure adding to the sense of solitude and reflection. The scene evokes a feeling of tranquility and isolation, capturing the beauty of a moment alone with nature.
Prompt
facial-expressions Daydreaming: Reflective, introspective ; A woman walking alone on a beach; eye-level; Single Person; vast, empty beach with crashing waves; cinematic
Characteristic
Shot : A woman walks away from the camera towards the horizon on a beach, with a second person walking further away in the distance. The sky is cloudy with some sun shining through. The beach is wet, with reflections of the sky and clouds in the water.
Aesthetic Score : 0.7
Mood : melancholy, contemplative, lonely
Quality
Entropy : 6.73
Noise : 104
Prompt Clip Score : 0.23
AI Evaluation
Likelihood of AI : 0.10
Image errors : No visible errors.
A Moment of Hope: Superman Contemplates the City
A lone figure in a Superman costume stands on a rooftop, gazing out over a sprawling cityscape. The dramatic interplay of light and shadow, coupled with the moving clouds overhead, creates a sense of heroic contemplation and hope. This image captures the essence of a superhero’s unwavering commitment to protecting the city below.
Prompt
facial-expressions Daydreaming: Empowered, triumphant ; A superhero soaring through the sky; high angle; Hero; dramatic cloudscape with city skyline in the distance; cinematic
Characteristic
Shot : A man in a Superman costume stands on a rooftop overlooking a cityscape, his back to the viewer.
Aesthetic Score : 0.7
Mood : heroic, powerful, contemplative
Quality
Entropy : 6.94
Noise : 100
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.20
Image errors : No visible artifacts or errors.
Conclusion
The analysis shows 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 0.25, which is below average. This suggests that the model didn’t accurately capture the intended camera position described in the prompt.
- Shot Analysis: The model scored 0.435, which is also below average. This indicates 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 0.11, which is within the very good range. This means that the generated image’s aesthetic closely matched the expected aesthetic described in the prompt.
Overall, the model seems to be struggling with accurately interpreting the camera position and scene description, but it managed to create an image with 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://leonardo.ai