AI's Struggle with Noir: A Look at Generative Models and Aesthetic Style with Dall-e-3
- 10 minutes read - 2084 wordsTable of Contents
Film noir, with its shadowy figures, stark contrasts, and gritty realism, has captivated audiences for decades. Its distinctive aesthetic, characterized by low-key lighting, dramatic angles, and a sense of urban decay, is instantly recognizable. But can AI capture the essence of this iconic style? This blog post explores the challenges of generating images with a specific aesthetic style, using film noir as an example. We’ll delve into the results of a generative AI model tasked with creating images based on film noir prompts, analyzing its strengths and weaknesses in capturing the desired mood and atmosphere.
Created with: dall-e-3
Lost in the Shadows: A Man’s Solitude in a Rainy Alley
A solitary figure, cloaked in a trench coat, stands amidst the glistening reflections of a rain-soaked alleyway. The dim glow of streetlights casts long shadows, amplifying the mood of darkness and melancholy. This evocative scene captures a moment of isolation and introspection, leaving the viewer to ponder the man’s story.
Prompt
Film noir: Gritty, determined, melancholic ; A lone detective; medium shot; Heroism; A dimly lit alleyway with rain pouring down, neon signs reflecting in puddles.; cinematic
Characteristic
Shot : A man in a trench coat stands in a rain-soaked city street at night, looking down. The street is wet and glistening with reflections of the street lights. There are blurred figures of other people in the background. The atmosphere is moody and melancholic. The image is well-lit and the composition is strong, creating a sense of depth and drama.
Aesthetic Score : 0.7
Mood : melancholy, mysterious, somber
Quality
Entropy : 6.81
Noise : 111
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image appears slightly blurry, but this could be intentional to create a moody atmosphere.
Lost in the Shadows: A Figure Races Through the Night Market
A single figure dashes through a dimly lit, bustling night market, their silhouette stark against the lanterns hanging overhead. The vanishing point at the end of the narrow alley adds a sense of mystery and suspense, leaving the figure’s destination unknown.
Prompt
Film noir: Suspenseful, mysterious, thrilling ; A shadowy figure escaping through a maze of narrow streets; long shot; Adventure; A bustling night market with flickering lanterns and crowded stalls.; cinematic
Characteristic
Shot : A shadowy, narrow alleyway in an Asian city. The alley is lit by hanging lanterns and there are many people walking along the alleyway, all facing towards a bright light at the end. The alley is lined with shops and stalls. A single figure in the center of the image is running towards the light.
Aesthetic Score : 0.7
Mood : mysterious, suspenseful, urban
Quality
Entropy : 5.84
Noise : 119
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image has some minor artifacts, such as the repeated pattern of the lanterns and the slightly blurred edges of the figures.
Four Aces and a Cloud of Smoke: The Big Win Awaits
A dimly lit casino, smoke swirling in the air, and a hand holding four aces - the ultimate poker hand. This image captures the thrill and suspense of a potential big win, leaving you wondering what happens next.
Prompt
Film noir: Intense, suspenseful, dangerous ; A gambler’s hand holding a pair of aces; close-up; Gaming; A smoky, dimly lit casino with flashing lights and the sound of slot machines.; cinematic
Characteristic
Shot : A hand holding a four of a kind aces poker hand in a smoky casino setting, with slot machines in the background and a table in the foreground
Aesthetic Score : 0.6
Mood : mysterious, suspenseful, dramatic
Quality
Entropy : 6.62
Noise : 91
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.30
Image errors : The image appears to have some noise and grain, which may be due to the low-light conditions or post-processing. There are also some slight artifacts around the edges of the slot machines.
Lost in the Rain: A Vintage Car’s Lonely Journey
A vintage car speeds down a deserted highway, swallowed by the rain and a desolate landscape. A distant motel beckons, promising a fleeting escape from the melancholic mood that hangs heavy in the air. This image captures a sense of isolation and mystery, leaving you wondering about the driver’s destination and the secrets hidden within the rain-soaked scenery.
Prompt
Film noir: Lonely, melancholic, atmospheric ; A vintage car speeding through a deserted highway; wide shot; Tourism; A desolate, rain-soaked landscape with a lone motel in the distance.; cinematic
Characteristic
Shot : A classic car driving down a lonely desert highway, with a motel in the distance, during a rainstorm.
Aesthetic Score : 0.7
Mood : melancholic, nostalgic, atmospheric
Quality
Entropy : 6.58
Noise : 112
Prompt Clip Score : 0.36
AI Evaluation
Likelihood of AI : 0.90
Image errors : The rain drops are not very realistic and the reflections in the road are a bit strange.
Lost in the Rain: A Cityscape of Melancholy
A woman gazes out of a train window, her reflection lost in the shimmering city lights reflected in the raindrops. The scene evokes a sense of melancholic nostalgia, with a touch of mystery in the dimly lit cityscape.
Prompt
Film noir: Nostalgic, melancholic, introspective ; A woman looking out of a train window at a passing cityscape; medium shot; Travel; A dark, rainy cityscape with towering buildings and flickering streetlights.; cinematic
Characteristic
Shot : A woman is looking out the window of a train, the train is moving through a rainy city. The city is dark and the buildings are tall and imposing.
Aesthetic Score : 0.7
Mood : melancholy, mysterious, brooding
Quality
Entropy : 6.31
Noise : 113
Prompt Clip Score : 0.35
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image has some noise, especially in the darker areas. The woman’s skin is a bit too smooth and her eyes are a bit too sharp. There are some unnatural looking light artifacts.
A Family’s Hope in the Gloom
A dimly lit room, illuminated by a single kerosene lamp, reveals a family gathered together. Their faces, etched with a mix of somberness and hope, tell a story of resilience in the face of hardship. The intimate setting and the vulnerability of the scene create a powerful sense of drama and tension.
Prompt
Film noir: Intimate, tense, claustrophobic ; A family huddled together in a dimly lit room; medium shot; Family; A cramped apartment with peeling paint and a flickering gas lamp.; cinematic
Characteristic
Shot : A group of people, likely a family, sitting in a dimly lit room. They are huddled together with a single oil lamp providing the only light. The room appears to be in a state of disrepair.
Aesthetic Score : 0.8
Mood : somber, contemplative, hope
Quality
Entropy : 6.65
Noise : 99
Prompt Clip Score : 0.25
AI Evaluation
Likelihood of AI : 0.20
Image errors : Some minor artifacts are present in the image. The lighting is a bit too harsh. The composition is a bit too crowded and cluttered.
Lost in the City Lights
A man in a dark suit, shrouded in shadow, contemplates the rain-soaked cityscape. The neon glow of a Chinese sign casts an ethereal light, adding to the mysterious and melancholic mood.
Prompt
Film noir: Mysterious, brooding, contemplative ; A detective’s silhouette against a window, cigarette smoke swirling around him; low-angle shot; Heroism; A rain-soaked city street with neon signs reflecting in the puddles.; cinematic
Characteristic
Shot : A man is smoking a cigarette by a window. It is raining outside and the city lights are reflected in the window.
Aesthetic Score : 0.7
Mood : melancholy, moody, mysterious
Quality
Entropy : 5.88
Noise : 92
Prompt Clip Score : 0.36
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image has some noise and grain. The colors are also a bit muted.
A Hand in the Shadows: Suspense and Mystery in a Smoky Alley
A gloved hand tightly grips a pistol in a dimly lit, smoke-filled alley. The scene is heavy with suspense, as a shadowy figure lurks in the background, adding to the mystery and intrigue. This gritty image evokes a sense of danger and anticipation, leaving the viewer wondering what will happen next.
Prompt
Film noir: Suspenseful, dangerous, threatening ; A close-up of a hand holding a gun, the barrel pointed at a shadowy figure; extreme close-up; Adventure; A dark, smoky back alley with a flickering streetlight.; cinematic
Characteristic
Shot : A close-up of a gloved hand holding a gun with a blurry figure of a man in the background, the scene is set in a dark alleyway with smoke
Aesthetic Score : 0.7
Mood : dark, suspenseful, dangerous
Quality
Entropy : 6.64
Noise : 81
Prompt Clip Score : 0.29
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image appears to have some slight artifacts and blurring around the edges of the gun and the background figure. The smoke looks slightly artificial.
The Spin of Fate: A Roulette Wheel Under the Casino Lights
A close-up shot captures the heart of the casino - a roulette wheel bathed in soft, mysterious light. The chips surrounding the wheel hint at the high stakes and the suspense of the upcoming spin. This image evokes a sense of drama and anticipation, drawing you into the heart of the action.
Prompt
Film noir: Intense, suspenseful, unpredictable ; A roulette wheel spinning, the ball landing on a single number; close-up; Gaming; A dimly lit casino with the sound of chips clinking and people whispering.; cinematic
Characteristic
Shot : A close-up shot of a roulette wheel in a casino setting. The wheel is in focus and appears to be spinning, with the ball visible in the center. The green felt table, with other roulette tables in the background, and various chips around the wheel add to the scene.
Aesthetic Score : 0.8
Mood : dramatic, suspenseful, luxurious
Quality
Entropy : 6.47
Noise : 107
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image has some slight artifacts around the edges of the roulette wheel, which may be due to the image being digitally enhanced. The lighting appears a bit artificial.
A Lone Figure Walks Towards Hope in the Storm
A solitary figure traverses a wooden pier, shrouded in mist and fog, towards a lighthouse casting its beam into a stormy sea. The scene evokes a sense of mystery, loneliness, and a glimmer of hope amidst the dramatic elements.
Prompt
Film noir: Lonely, melancholic, atmospheric ; A lone figure walking down a deserted pier, the ocean waves crashing against the pilings; long shot; Travel; A foggy, desolate coastline with a lighthouse in the distance.; cinematic
Characteristic
Shot : A lone figure walks along a pier towards a lighthouse, shrouded in mist and stormy seas.
Aesthetic Score : 0.7
Mood : melancholic, mysterious, hopeful
Quality
Entropy : 6.68
Noise : 100
Prompt Clip Score : 0.34
AI Evaluation
Likelihood of AI : 0.50
Image errors : The image contains slight blurriness in areas such as the water and the figure’s silhouette, which might be due to deliberate artistic choices or post-processing.
Conclusion
The results show that the generative AI model performed well in terms of camera position and shot analysis, but struggled with aesthetic analysis.
Here’s a breakdown:
- Camera Position: The model scored 0.35, which is below the “good” range of 0.5 to 0.75. This suggests that the model didn’t always accurately translate the intended camera positions from the prompt into the generated image.
- Shot Analysis: The model scored 0.43, also below the “good” range. This indicates that the model had some difficulty understanding the scene described in the prompt and translating it into a visually coherent shot.
- Aesthetic Analysis: The model scored 0.29, which is significantly below the “very good” range of -0.2 to 0.1. This means that the generated image’s aesthetic deviated considerably from the expected aesthetic based on the prompt.
Overall, the model shows promise in understanding camera positions and shot composition, but needs improvement in capturing the desired aesthetic.
Sources:
- https://heartofnoir.com/knowing-noir/aesthetic-of-noir/
- https://www.yellowbrick.co/blog/film/maximizing-the-visual-impact-unveiling-the-art-of-film-aesthetics
- https://www.questjournals.org/jrhss/papers/vol10-issue8/1008255260.pdf
- https://www.jstor.org/stable/3331672
- https://www.cinepoetics.fu-berlin.de/activities/workshops/2020-12-ws/index.html
- https://resource.download.wjec.co.uk/vtc/2016-17/16-17_1-22/eng/Part%201%20What%20is%20Aesthetics.pdf
- https://openai.com/index/dall-e-3/