AI Captures Pride's Joy, But Struggles with Camera Angles with Flux-dev
- 9 minutes read - 1777 wordsTable of Contents
The world of AI is constantly evolving, and its ability to generate realistic and evocative images is becoming increasingly impressive. In this experiment, we tasked an AI model with creating images based on various Pride-themed scenes. The model demonstrated a strong understanding of the emotional and aesthetic aspects of Pride, generating images that captured the joy, celebration, and sense of community associated with these events. However, the model struggled with accurately representing camera angles, highlighting a key area for improvement in AI image generation.
Created with: flux-dev
Determined Message Sparks Curiosity in Crowded City
A lone figure stands amidst a bustling crowd, holding a sign that reads ‘Determined, haveful with is TTA’. The message, shrouded in mystery, draws attention and sparks intrigue. The scene exudes a sense of seriousness and hope, leaving viewers wondering about the meaning behind the cryptic words.
Prompt
facial-expressions Pride: Determined, hopeful, powerful ; A person holding a sign with a message of acceptance; eye-level; Single Persons; A crowd of people at a Pride protest; cinematic
Characteristic
Shot : A person holding a sign with the words ‘Determined, haveful with is TTA’ in a crowd of people.
Aesthetic Score : 0.4
Mood : determined, hopeful, optimistic
Quality
Entropy : 6.70
Noise : 83
Prompt Clip Score : 0.21
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is slightly blurry, and the colors are a bit washed out.
Lost in Thought: A Moment of Wonder in the Crowd
A close-up portrait captures a woman’s contemplative gaze, her head tilted upwards as if lost in a dream. The soft lighting and blurred background create a dreamy atmosphere, leaving the viewer to wonder what she is seeing and feeling.
Prompt
facial-expressions Pride: Awe, inspiration, hope ; A person looking out at a Pride parade with a sense of wonder; eye-level; Single Persons; A vibrant parade with colorful floats and music; cinematic
Characteristic
Shot : Close-up portrait of a woman looking upwards, likely in a crowded outdoor setting. Her expression is contemplative and peaceful.
Aesthetic Score : 0.7
Mood : dreamy, contemplative, serene
Quality
Entropy : 6.68
Noise : 63
Prompt Clip Score : 0.18
AI Evaluation
Likelihood of AI : 0.20
Image errors : No noticeable artifacts or errors in the image.
Silhouettes and Spotlight: A Night of Energetic Dance
A woman in a white shirt takes center stage, her silhouette illuminated by vibrant red and blue lights. The energy of the party is palpable, with the woman’s movements capturing the fun and lively atmosphere. The dramatic lighting adds a touch of mystery, making this a captivating scene.
Prompt
facial-expressions Pride: Joyful, carefree, celebratory ; A group of people dancing in a club; eye-level; Normal People; A brightly lit dance floor with rainbow lights; cinematic
Characteristic
Shot : A group of people dancing at a club or party. The scene is dimly lit with colored lights. A woman with sunglasses is dancing in the foreground.
Aesthetic Score : 0.6
Mood : energetic, fun, vibrant
Quality
Entropy : 6.36
Noise : 54
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.10
Image errors : There are some minor artifacts in the image, particularly in the background. The lighting is also somewhat uneven.
Pride Celebration: A Moment of Joy and Pride
A young woman, radiating joy and pride, celebrates at a vibrant Pride parade. Her raised arms and the colorful rainbow flag in the background capture the spirit of the event, creating a sense of excitement and celebration.
Prompt
facial-expressions Pride: Empowered, defiant, hopeful ; A person holding a rainbow flag high; eye-level; Single Persons; A crowd of people at a Pride rally; cinematic
Characteristic
Shot : A woman in a white tank top is holding up her arms in the air, celebrating a pride parade. She is standing in the street, with a rainbow flag in the background. The city backdrop and other people in the scene create a festive atmosphere.
Aesthetic Score : 0.7
Mood : joyful, celebratory, hopeful
Quality
Entropy : 6.84
Noise : 59
Prompt Clip Score : 0.22
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image is slightly blurry and there are some artifacts in the background.
Laughter and Freedom: A Moment of Joy Captured
This image captures a young woman with blonde curly hair, eyes closed in laughter, radiating pure joy. The blurred background, featuring people and a rainbow flag, adds to the feeling of carefree happiness and freedom. The genuine expression and the soft focus create a sense of warmth and lightheartedness.
Prompt
facial-expressions Pride: Joyful, confident, celebratory ; A single person; eye-level; Single Persons; A bustling Pride parade with rainbow flags and confetti; cinematic
Characteristic
Shot : A woman laughing at a pride parade
Aesthetic Score : 0.7
Mood : joyful, celebratory, carefree
Quality
Entropy : 6.57
Noise : 65
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.20
Image errors : Slight blur in the background.
Lost in a World of Color: A Moment of Joy and Wonder
A person is captivated by a computer screen, their eyes drawn to a cartoon character adorned in a vibrant rainbow shirt. The scene radiates happiness, playfulness, and hope, leaving the viewer with a sense of wonder and curiosity.
Prompt
facial-expressions Pride: Creative, playful, inclusive ; A gamer creating a rainbow-themed character in a video game; eye-level; Gamer; A computer screen with a character creation menu; cinematic
Characteristic
Shot : A person is looking at a computer screen with a 3D animated character on it. The character is wearing a rainbow shirt and smiling
Aesthetic Score : 0.7
Mood : playful, happy, hopeful
Quality
Entropy : 6.75
Noise : 58
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.80
Image errors : No major errors
Sunset Romance: A Couple’s Walk Towards Happiness
A heartwarming scene of a couple strolling hand-in-hand towards the setting sun. The man gazes lovingly at the woman, who beams back at him, radiating joy and hope. The warm, golden light of the sunset creates a romantic and idyllic backdrop for their shared moment.
Prompt
facial-expressions Pride: Loving, peaceful, accepting ; A couple holding hands and walking down the street; eye-level; Normal People; A quiet, residential street with rainbow flags on display; cinematic
Characteristic
Shot : A couple is walking down the street, hand in hand, with a sunset in the background. There are some buildings in the background, as well as some other people in the distance.
Aesthetic Score : 0.6
Mood : romantic, happy, carefree
Quality
Entropy : 6.66
Noise : 84
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image has a slightly blurry background, and the lighting is not perfectly balanced. The colors are a bit washed out.
Rainbow Smiles: Capturing the Joy of Pride
A man beams with happiness, his vibrant rainbow cape a beacon of pride against a backdrop of celebration. The blurry background suggests the energy and movement of a parade, capturing the spirit of hope and joy that defines this momentous occasion.
Prompt
facial-expressions Pride: Powerful, inspiring, hopeful ; A superhero in a rainbow costume; eye-level; Heroes; A cityscape with a Pride flag flying in the background; cinematic
Characteristic
Shot : A man in a rainbow shirt stands in front of a blurred background of other people and a rainbow flag.
Aesthetic Score : 0.7
Mood : joyful, positive, proud
Quality
Entropy : 6.80
Noise : 80
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image has slight blurriness, especially in the background.
Laughter and Light: Capturing the Joy of the Moment
Two young women share a genuine laugh, their joy radiating through the soft lighting and warm atmosphere. The blurry background hints at a celebratory setting, making this image a perfect snapshot of pure happiness.
Prompt
facial-expressions Pride: Joyful, celebratory, inclusive ; A group of friends celebrating at a Pride party; eye-level; Normal People; A brightly decorated room with rainbow decorations; cinematic
Characteristic
Shot : Three young women are laughing together in a brightly lit room with blurred lights in the background. The image is likely taken at a party or celebration.
Aesthetic Score : 0.8
Mood : happy, joyful, celebratory
Quality
Entropy : 6.42
Noise : 62
Prompt Clip Score : 0.26
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image is slightly overexposed, resulting in a washed-out look, especially in the background. There’s also some noise in the shadows.
Immersed in the Game: A Gamer’s World Lit in Pink and Blue
This image captures the focused intensity of a gamer, lost in the digital world. The vibrant pink and blue lighting adds a playful touch, highlighting the immersive experience of gaming.
Prompt
facial-expressions Pride: Fun, playful, inclusive ; A gamer playing a video game with rainbow-themed characters; eye-level; Gamer; A brightly lit gaming room with posters of LGBTQ+ characters; cinematic
Characteristic
Shot : A young man is sitting at a desk in front of a computer, wearing headphones and smiling. He is looking at the screen, which is showing a group of people. There is a lamp in the background and some artwork on the wall.
Aesthetic Score : 0.6
Mood : joyful, focused, playful
Quality
Entropy : 6.76
Noise : 61
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is a bit blurry. There is some noise in the background. The colors are a bit oversaturated.
Conclusion
The analysis shows 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 0.2, which is considered below average. This suggests that the generated image didn’t accurately reflect the camera position described in the prompt.
- Shot Analysis: The model scored 0.67, which is considered good. This indicates that the model was able to understand the scene described in the prompt and create an image that reflects it well.
- Aesthetic Analysis: The model scored 0.07, which is considered very good. This means that the generated image closely matched the expected aesthetic style.
Overall, the model seems to be better at understanding the scene and achieving the desired aesthetic, but it needs improvement in accurately capturing the 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://fal.ai/models/fal-ai/flux/dev/api