AI Struggles to Capture Gothic Aesthetics with Imagen-v2
- 10 minutes read - 2031 wordsTable of Contents
The dramatic style, often associated with gothic aesthetics, is characterized by its use of strong contrasts, dramatic lighting, and evocative imagery. It’s a style that has been used in countless works of art, literature, and film, from the paintings of Caspar David Friedrich to the novels of Mary Shelley to the films of Tim Burton. But can AI truly understand and replicate this style? In this blog post, we explore the challenges of using AI to generate images with specific aesthetic styles, focusing on the example of gothic aesthetics. We analyze the results of a test where an AI model was tasked with creating images based on various gothic-inspired scenes. The results highlight the model’s strengths in understanding scene and camera position, but its weaknesses in capturing the desired aesthetic. We delve into the reasons behind this discrepancy and discuss the implications for the future of AI-generated art.
Created with: imagen-v2
Silhouetted Against the Red Sky: A Lone Figure in a Desolate Landscape
A solitary figure stands amidst a ruined tower and a desolate landscape, bathed in the eerie glow of a fiery sunset. The scene evokes a sense of melancholy and isolation, with the dramatic red sky adding to the mystery and foreboding atmosphere.
Prompt
Gothic: Epic and melancholic ; A lone knight, silhouetted against a blood-red sunset; wide shot; Heroism; A crumbling castle on a windswept cliff; cinematic
Characteristic
Shot : A lone figure stands on a red hill overlooking a sea as the sun sets. A crumbling tower sits on a peak in the distance, shrouded in the red glow of sunset.
Aesthetic Score : 0.7
Mood : lonely, mysterious, dramatic
Quality
Entropy : 6.59
Noise : 79
Prompt Clip Score : 0.37
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image appears to have been heavily edited, perhaps with an AI filter, which has caused an unnatural red hue and some blurring in the distance.
Secrets Unveiled in Candlelight
A close-up shot of a dimly lit room, bathed in the warm glow of a single candle. A weathered map lies open on a wooden surface, inviting exploration. Old books line the shelves in the background, whispering tales of forgotten times. This image evokes a sense of mystery, antiquity, and nostalgia, drawing you into a world of secrets waiting to be discovered.
Prompt
Gothic: Intriguing and mysterious ; A weathered map, illuminated by flickering candlelight; close-up; Adventure; A dusty, cobweb-filled library; cinematic
Characteristic
Shot : A dimly lit desk with a lit candle, a map, a magnifying glass, a quill, and a row of books in the background.
Aesthetic Score : 0.7
Mood : mysterious, antique, suspenseful
Quality
Entropy : 6.49
Noise : 67
Prompt Clip Score : 0.31
AI Evaluation
Likelihood of AI : 0.40
Image errors : There’s a slight blur in the foreground map and the candle flame. The lighting seems unnatural.
A Hand Reaches Out From the Screen, But What Lies Beyond?
A chilling image of a skeletal hand emerging from a shattered television screen, hinting at a world beyond the digital realm. The blurred background and nearby video game controller add to the eerie atmosphere, leaving viewers to ponder the mystery behind this unsettling scene.
Prompt
Gothic: Eerie and unsettling ; A skeletal hand reaching out from a cracked screen; close-up; Gaming; A dimly lit room filled with gaming consoles and flickering monitors; cinematic
Characteristic
Shot : A skeletal hand reaches out from a broken screen, a video game controller lies nearby on a dark surface
Aesthetic Score : 0.7
Mood : dark, eerie, unsettling
Quality
Entropy : 6.03
Noise : 95
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image is slightly over-sharpened, which creates a halo effect around the edges of the objects.
A Solitary Figure Walks Towards a Gothic Mystery
A single person traverses a rain-slicked cobblestone street, their path leading towards a towering gothic church. The overcast sky and the lone figure create an atmosphere of mystery and somber reflection.
Prompt
Gothic: Awe-inspiring and melancholic ; A lone figure standing on a cobblestone street, gazing at a towering cathedral; medium shot; Tourism; A misty, rain-soaked European city; cinematic
Characteristic
Shot : A solitary figure walks down a cobblestone street towards a gothic cathedral, the street is wet and the sky is grey, suggesting a rainy day
Aesthetic Score : 0.7
Mood : gloomy, atmospheric, mysterious
Quality
Entropy : 6.73
Noise : 101
Prompt Clip Score : 0.37
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image has some minor artifacts, particularly around the edges of the buildings. The cobblestones look slightly blurry and lack detail, suggesting potential over-sharpening. The sky appears slightly unnatural and lacking texture
Iron Horse Against the Storm
A powerful steam locomotive bursts from a mountain station, its plume of smoke a defiant gesture against the brooding, dark clouds overhead. The scene evokes a sense of drama, nostalgia, and the raw power of the industrial age.
Prompt
Gothic: Dramatic and suspenseful ; A vintage train hurtling through a dark, stormy landscape; long shot; Travel; A desolate, gothic-inspired train station; cinematic
Characteristic
Shot : A steam locomotive is leaving a station, it’s covered in smoke and there are mountains in the background
Aesthetic Score : 0.75
Mood : dramatic, nostalgic, mysterious
Quality
Entropy : 6.81
Noise : 92
Prompt Clip Score : 0.34
AI Evaluation
Likelihood of AI : 0.80
Image errors : The smoke and the steam engine look slightly too smooth, possible AI generated
Shadows and Secrets: A Gothic Gathering by the Fire
A trio of figures huddle in the warm glow of a fireplace, their faces obscured by shadows. The gothic setting, with its stained glass windows and dim lighting, creates an atmosphere of mystery and intrigue. This captivating scene is a study in light and shadow, highlighting the figures while leaving the details of the room shrouded in darkness.
Prompt
Gothic: Warm and intimate ; A family huddled around a fireplace, shadows dancing on the walls; medium shot; Family; A grand, gothic-style mansion with stained glass windows; cinematic
Characteristic
Shot : Three figures are huddled together near a fireplace in a dimly lit room. The room has two large windows and a fireplace. There are many candles and other items on the mantle above the fireplace.
Aesthetic Score : 0.7
Mood : mysterious, somber, enchanting
Quality
Entropy : 6.69
Noise : 92
Prompt Clip Score : 0.32
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image has a slightly blurry quality, and there is some pixelation around the edges of the figures. Some of the details are indistinct, and the image appears somewhat artificially rendered.
Raven’s Watch: A Medieval Mystery Unfolds
A solitary raven perches on a weathered gargoyle, its sharp gaze fixed on the distant, looming castle. The scene evokes a sense of mystery and foreboding, hinting at secrets hidden within the ancient walls. The depth of field draws the viewer’s attention to the raven, emphasizing its role as a watchful observer in this medieval world.
Prompt
Gothic: Mysterious and ominous ; A lone raven perched on a gargoyle, overlooking a bustling city; close-up; Heroism; A gothic cathedral with intricate carvings and stained glass; cinematic
Characteristic
Shot : A crow perched on a stone ledge with a large building with gothic architecture in the background
Aesthetic Score : 0.8
Mood : dark, mysterious, ominous
Quality
Entropy : 6.95
Noise : 70
Prompt Clip Score : 0.35
AI Evaluation
Likelihood of AI : 0.20
Image errors : There are no noticeable artifacts or errors in the image.
A Candle’s Glow Unveils Hidden Treasures
A single candle casts a warm, inviting glow upon a treasure chest overflowing with gems and riches. The scene is shrouded in mystery, with spiderwebs clinging to the background, hinting at a forgotten past. The anticipation of discovery hangs heavy in the air, beckoning you to explore the secrets within.
Prompt
Gothic: Excitement and danger ; A treasure chest overflowing with gold and jewels, illuminated by a single candle; close-up; Adventure; A dark, damp dungeon with cobwebs and chains; cinematic
Characteristic
Shot : A wooden chest, half-open, with a lit candle inside. It is surrounded by various jewels and a web-like structure in the background.
Aesthetic Score : 0.7
Mood : mysterious, adventurous, nostalgic
Quality
Entropy : 6.38
Noise : 88
Prompt Clip Score : 0.33
AI Evaluation
Likelihood of AI : 0.30
Image errors : There are minor blurriness and graininess in the image, particularly in the background. The lighting and shadows also appear somewhat artificial.
The Grim Reaper’s Army: A Skeletal Horde Descends
A lone skeletal warrior, clad in dark armor, stands defiant amidst a vast, menacing horde of his brethren. The scene is shrouded in a thick, oppressive mist, creating an atmosphere of impending doom and apocalyptic despair. The skeletal warrior’s pose, a mixture of power and grim determination, adds to the dramatic effect, leaving viewers with a chilling sense of foreboding.
Prompt
Gothic: Grim and triumphant ; A player’s avatar, a skeletal warrior, standing amidst a graveyard of fallen enemies; medium shot; Gaming; A dark and eerie virtual world with gothic architecture; cinematic
Characteristic
Shot : A skeletal figure clad in dark armor stands amidst a chaotic scene. A horde of figures can be seen in the foreground and background, suggesting a battle or conflict. The skeletal figure is holding a large sword, further emphasizing the sense of threat and power.
Aesthetic Score : 0.7
Mood : dark, intense, ominous
Quality
Entropy : 6.93
Noise : 61
Prompt Clip Score : 0.34
AI Evaluation
Likelihood of AI : 0.80
Image errors : The image contains some visual artifacts, particularly in the background and on the edges of the skeletal figure. This suggests it might be AI-generated. The rendering of certain elements, particularly the sword, also appears slightly unnatural.
A Solitary Figure and a Majestic Castle in a Mystical Landscape
A lone figure stands on a cliff, gazing out at a vast, desolate landscape. In the distance, a towering castle rises from a rocky mountain, casting a long shadow over the scene. The image evokes a sense of mystery, loneliness, and grandeur, with the dramatic composition highlighting the isolation of the figure and the imposing scale of the castle.
Prompt
Gothic: Awe-inspiring and melancholic ; A lone traveler standing at the edge of a cliff, gazing at a sprawling, gothic-inspired city; wide shot; Travel; A mountainous landscape with dramatic clouds and a stormy sky; cinematic
Characteristic
Shot : A lone figure stands on a cliff overlooking a vast, desolate landscape. In the distance, a towering, gothic castle rises from a mountain peak.
Aesthetic Score : 0.7
Mood : mysterious, epic, lonely
Quality
Entropy : 6.59
Noise : 94
Prompt Clip Score : 0.34
AI Evaluation
Likelihood of AI : 0.90
Image errors : The image has some minor artifacts, such as the jagged edges of the castle and the blurry areas in the distance.
Conclusion
The results show 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 0.4, indicating a fair performance. This means the camera position in the generated image was somewhat different from what was requested in the prompt. While not excellent, it’s still within the acceptable range.
- Shot Analysis: The model scored 0.52, indicating a good performance. This means the generated image captured the scene in a way that was fairly close to what was described in the prompt.
- Aesthetic Analysis: The model scored 0.3, indicating a poor performance. This means the generated image’s aesthetic was significantly different from what was expected based on the prompt.
Overall, the model seems to be better at understanding the scene and camera position than it is at 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://deepmind.google/technologies/imagen-2/