AI's Artistic Eye: A Look at Generative Models and Aesthetic Style with Titan-g1
- 9 minutes read - 1835 wordsTable of Contents
The world of visual storytelling is constantly evolving, and AI is playing an increasingly significant role in shaping its future. One fascinating aspect of this evolution is the ability of AI to generate images that embody specific aesthetic styles. This blog post explores a case study where we analyze the performance of an AI model in capturing the essence of a particular aesthetic style. We’ll delve into the model’s strengths and limitations, highlighting its success in replicating shot composition and aesthetic, while also examining its challenges in accurately capturing camera positions. By understanding these nuances, we can gain valuable insights into the potential and limitations of AI in the realm of visual storytelling.
Created with: titan-g1
Silhouette of Solitude: A Figure Walks Towards the Setting Sun
A lone figure traverses a vast, white sandy desert, their silhouette stark against the glowing sunset. The scene evokes a sense of solitude, contemplation, and peace, with the dramatic effect of the figure’s mystery and the desert’s isolation adding intrigue.
Prompt
French New Wave: epic, melancholic ; A lone figure, silhouetted against a setting sun; long shot; heroism; a vast, empty desert landscape; cinematic
Characteristic
Shot : A lone figure walks across a vast desert landscape during sunset. The sun is setting in the distance, casting a warm glow over the scene. The sand dunes are gently rolling, and the sky is a soft, pale blue.
Aesthetic Score : 0.7
Mood : tranquil, serene, contemplative
Quality
Entropy : 6.40
Noise : 101
Prompt Clip Score : 0.22
AI Evaluation
Likelihood of AI : 0.20
Image errors : The image has noticeable grain, which detracts from its sharpness. The colors also appear slightly faded.
The Call of the Open Road: A Hand Points the Way
A nostalgic and adventurous scene unfolds as a hand points towards a map, surrounded by travel essentials like a compass. The anticipation and excitement for the journey ahead are palpable, inviting you to embark on your own exploration.
Prompt
French New Wave: intriguing, suspenseful ; A close-up of a weathered map, with a finger tracing a route; medium shot; adventure; a cluttered, dimly lit room; cinematic
Characteristic
Shot : A hand points to a location on a folded map, with a compass or other device in the background, giving a sense of travel and exploration.
Aesthetic Score : 0.5
Mood : mysterious, adventurous, contemplative
Quality
Entropy : 6.65
Noise : 101
Prompt Clip Score : 0.21
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image is slightly blurry and the colors are slightly washed out.
Reliving the Arcade Dream: A Hand on the Joystick
A close-up shot captures the thrill of classic arcade gaming. The hand gripping the joystick, the neon glow of the screen, and the blurred gameplay all evoke a sense of nostalgia and playful excitement. This image transports you back to the golden age of arcades, where every game was an adventure.
Prompt
French New Wave: intense, energetic ; A hand holding a joystick, fingers moving rapidly; close-up; gaming; a neon-lit arcade with flashing screens; cinematic
Characteristic
Shot : Close up of a person’s hand playing an arcade game. The game is blurry in the background, but it looks like it could be a racing game.
Aesthetic Score : 0.5
Mood : nostalgic, retro, focused
Quality
Entropy : 6.02
Noise : 101
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image has a bit of blur, but it is not too noticeable.
Parisian Dreams: A Woman Gazes at the Eiffel Tower
A wistful moment captured on a Parisian street. A woman in a blue jacket stands, her gaze fixed on the iconic Eiffel Tower in the distance. The scene evokes a sense of longing and wonder, capturing the romantic spirit of the City of Lights.
Prompt
French New Wave: romantic, nostalgic ; A young woman, her face filled with wonder, gazing at the Eiffel Tower; medium shot; tourism; a bustling Parisian street; cinematic
Characteristic
Shot : A woman is looking at the Eiffel Tower in Paris, she is wearing a dark jacket, the scene is in a Parisian street with buildings on both sides
Aesthetic Score : 0.7
Mood : romantic, dreamy, nostalgic
Quality
Entropy : 6.88
Noise : 105
Prompt Clip Score : 0.27
AI Evaluation
Likelihood of AI : 0.20
Image errors : There are some minor artifacts in the image, particularly in the background. The image appears slightly over-exposed
Blurred Beauty: A Train Ride Through Fields of Gold
A wistful journey captured through a train window, showcasing a vibrant field of yellow flowers and a lush green hedge. The fast-moving train creates a mesmerizing motion blur, adding a sense of dynamism and nostalgia to the tranquil scene.
Prompt
French New Wave: reflective, contemplative ; A train speeding through a countryside landscape, with a lone figure looking out the window; long shot; travel; a vibrant, sun-drenched field; cinematic
Characteristic
Shot : A view from a train window, looking out at a field of yellow flowers, the train is moving quickly, creating a motion blur effect in the image
Aesthetic Score : 0.6
Mood : tranquil, nostalgic, journey
Quality
Entropy : 6.70
Noise : 105
Prompt Clip Score : 0.28
AI Evaluation
Likelihood of AI : 0.20
Image errors : No noticeable artifacts or errors
The Joy of Family Gatherings
A heartwarming photo capturing a family sharing a meal, filled with laughter and genuine connection. The warm and relaxed atmosphere radiates happiness, showcasing the beauty of family moments.
Prompt
French New Wave: intimate, heartwarming ; A family gathered around a table, sharing a meal, with laughter and conversation; medium shot; family; a warm, inviting kitchen; cinematic
Characteristic
Shot : A family gathering around a table set for a meal. The woman in the center of the image is smiling and serving food while the other people at the table look at her. There is a window with white curtains in the background.
Aesthetic Score : 0.7
Mood : joyful, heartwarming, casual
Quality
Entropy : 6.94
Noise : 108
Prompt Clip Score : 0.34
AI Evaluation
Likelihood of AI : 0.10
Image errors : The image has a slight graininess and a few artifacts around the edges. The colors are a bit muted.
Lost in the Parisian Labyrinth
A young man navigates the bustling streets of Paris, his hurried pace hinting at a secret mission. The vintage atmosphere and narrow alleyways create a sense of mystery and intrigue, leaving you wondering what secrets lie ahead.
Prompt
French New Wave: urgent, dramatic ; A young man, his face etched with determination, running through a crowded marketplace; medium shot; heroism; a chaotic, bustling market; cinematic
Characteristic
Shot : A man in a suit runs through a crowded Parisian street, captured in a candid snapshot.
Aesthetic Score : 0.6
Mood : mysterious, urban, tense
Quality
Entropy : 6.55
Noise : 104
Prompt Clip Score : 0.32
AI Evaluation
Likelihood of AI : 0.10
Image errors : Some minor noise and grain are present, which is expected for a photo of this age.
The Compass: A Timeless Guide
A close-up shot of a classic compass, its needle pointing north, evokes a sense of minimal elegance and timeless precision. The focus on detail highlights the instrument’s enduring importance in navigation and exploration.
Prompt
French New Wave: mysterious, suspenseful ; A close-up of a compass needle spinning, pointing towards an unknown destination; close-up; adventure; a dimly lit, mysterious room; cinematic
Characteristic
Shot : Close-up shot of a compass with a golden needle pointing north
Aesthetic Score : 0.7
Mood : classic, elegant, adventurous
Quality
Entropy : 6.38
Noise : 111
Prompt Clip Score : 0.18
AI Evaluation
Likelihood of AI : 0.20
Image errors : Slight noise in the image, especially in the background
Intrigued by the Screen: A Moment of Focused Curiosity
Four young faces, illuminated by the soft glow of a laptop screen, are caught in a moment of intense focus. The dim lighting and close-up framing create a sense of mystery and draw the viewer into their shared experience.
Prompt
French New Wave: intense, focused ; A group of friends huddled around a computer screen, their faces illuminated by the glow; medium shot; gaming; a dimly lit, cluttered room; cinematic
Characteristic
Shot : Four young people are gathered around a computer screen, looking intently at something on the screen. The scene is dimly lit, with only the glow of the computer screen illuminating their faces. The composition of the image is good, with a clear sense of depth and perspective.
Aesthetic Score : 0.6
Mood : intense, focused, suspenseful
Quality
Entropy : 6.29
Noise : 110
Prompt Clip Score : 0.24
AI Evaluation
Likelihood of AI : 0.10
Image errors : A bit of grain or film noise is present in the image, likely a stylistic choice.
Lost in Love, Amidst European Charm
A couple strolls hand-in-hand down a picturesque European street, their intimacy highlighted against the backdrop of charming old buildings. The scene evokes a sense of romance, quaintness, and nostalgia, capturing the essence of a timeless love story.
Prompt
French New Wave: romantic, nostalgic ; A couple walking hand-in-hand along a cobblestone street, their silhouettes framed by the setting sun; long shot; tourism; a romantic, picturesque town; cinematic
Characteristic
Shot : A couple walks down a narrow street lined with old stone buildings. The sun is shining brightly, and the buildings cast long shadows.
Aesthetic Score : 0.7
Mood : romantic, nostalgic, quaint
Quality
Entropy : 6.65
Noise : 108
Prompt Clip Score : 0.28
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 terms of camera position and shot analysis, but struggled with aesthetic analysis.
Here’s a breakdown:
- Camera Position: The model scored 0.4, which is considered okay. This means the generated image’s camera position was somewhat different from what was requested in the prompt.
- Shot Analysis: The model scored 0.525, which is considered good. This indicates the generated image’s shot composition was fairly close to what was described in the prompt.
- Aesthetic Analysis: The model scored 0.12, which is considered very good. This means the generated image’s aesthetic was very close to the expected aesthetic.
Overall, the model seems to be better at understanding and implementing shot composition and aesthetic style than it is at accurately capturing camera position.
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://docs.aws.amazon.com/bedrock/latest/userguide/titan-image-models.html