A guide to AI x Street Photography
May 2025
May 2025
This photo exhibition was created for the FILM 206 class in Spring 2025. Centered around the theme of street photography, this project presents a concise series of street scenes. I utilized AI tools throughout the creative process—experimenting with their ability to assist, enhance, and even reinterpret the act of photographic observation.
1. Photo to Text to Photo
A photograph taken by a camera is first given to an AI to generate a textual description, and then this description is fed into an image-generation AI to produce a new picture. This process shows that AI can grasp the main elements of the scene and relatively accurately translate them into text, eventually generating an image with the same basic components.
However, it is clear that the AI struggles to capture finer details and is prone to errors when dealing with objects whose forms are vague or difficult to distinguish.
My Photo
AI generated Text Based on My Photo
AI Generated Photo Based on the Text
Photographed at eye level with a centered, head-on composition, the image captures a single vintage chair placed directly against a graffitied urban wall. The framing is tight, with the chair occupying the central third of the frame, creating a symmetrical and grounded visual weight. The ornate baroque-style wooden frame of the chair is painted in a faded mint green, its curved legs and carved details weathered by time. The upholstery is a pale yellow-beige fabric, visibly frayed and torn at the seat, with stuffing peeking through.
Behind it, a layered wall of brick and wood is coated in multicolored street tags-pink, turquoise, black, and violet-giving a sense of gritty, spontaneous expression. The sidewalk foreground is slightly scuffed concrete, with subtle marks and dirt, adding realism. Natural diffused light creates soft shadows and even contrast, emphasizing texture and decay.
Shot from an elevated, top-down perspective, this photograph captures a solitary figure seated at the base of a narrow concrete staircase between two red-brick buildings. The framing is symmetrical, with black metal railings on both sides guiding the viewer's eye downward toward the hooded person, who sits hunched forward, their back to the camera. The subject wears a dark gray hoodie, blending into the muted tones of the environment.
A black backpack lies on the ground nearby, partially obscured by shadow. Beyond the stairs, a patch of sparse grass meets a stained concrete path, leading to a large blue industrial dumpster in the upper-right corner of the frame. The lighting is natural but dim, possibly early evening, adding a subdued, slightly melancholic tone to the image. The composition evokes a sense of isolation, urban stillness, and unnoticed human presence in a back alley or service area.
A downward-facing shot taken from an elevated position captures an industrial alleyway flanked by tall, worn buildings-bare concrete on the left and weathered red brick on the right. The composition leads the viewer's eye down a sloped driveway lined with parked cars, descending toward a loading zone crowded with delivery trucks and vans. The camera angle emphasizes vertical lines and depth, creating a tunnel-like perspective. A few people are scattered near the trucks, mid-action-some walking, some standing. Ventilation pipes, ducts, and exposed infrastructure line the right-side brick wall, giving a gritty, utilitarian feel. A green pedestrian walkway crosses overhead, partially obscuring the midground. The color palette is muted and urban-concrete gray, brick red, and faded metal. The image captures a behind-the-scenes moment of city life, with a distant, observational tone typical of street photography.
2. Photo to Text to Photo (Abstraction)
My Photo
AI Generated Photo
When confronted with highly abstract or difficult-to-discern photographs, the AI’s analytical abilities become notably limited. The image it generates can only rely on a few abstract keywords to establish a general tone, but the final result often differs significantly from the original photo.
Due to the vagueness and limited information, the AI often amplifies certain features mentioned in the description, resulting in generated images with an exaggerated style.
3. AI Nude / Deepfake
My Photo
AI Generated
Since its invention, AI image manipulation technology has been widely used to create pornographic content—a pattern that mirrors the early histories of photography and film. Using AI to "undress" a person is remarkably easy.
Ironically, because AI training is deeply influenced by mainstream Western values, the resulting images often depict idealized body shapes and exaggerated sexual characteristics.
4. Decline of Text
Original Photo
This is a highly detailed, high-resolution photograph captured from a slightly elevated, diagonal angle, showing a sunny winter afternoon in a busy urban environment. In the foreground, several bright yellow bollards with red reflective stripes stand on a clean concrete sidewalk, some of them marked with black graffiti. A Black man, bald and wearing glasses, dressed in a vivid red jacket over a dark shirt, walks carefully through the gaps between the bollards, looking downward. Behind him, a bike lane with large white bicycle symbols runs alongside a line of heavy traffic, including white and black SUVs and a silver hatchback. On the left side, leafless winter trees cast complex shadows onto the street and sidewalk. Behind the trees, a brick building features a colorful mural with oversized, cartoon-like fruits, contrasting with the muted tones of the concrete surroundings. On the right, a plain grey brick wall and a wooden utility pole bearing a "Bike Lane" sign frame the scene. The sharp sunlight creates strong, crisp shadows and vivid contrast across the image, highlighting the textures of the cityscape in a bright, almost hyper-realistic manner.
A high-resolution photo taken from a slightly elevated, diagonal angle. A bald Black man in a vivid red jacket walks between bright yellow bollards with red reflective stripes, some marked with graffiti. Behind him, a bike lane with white bicycle symbols runs beside a traffic jam of SUVs. Leafless trees and a colorful fruit mural appear in the background. Sharp winter sunlight casts strong shadows across the scene.
A high-res photo of a bald Black man in a red jacket walking between yellow bollards under bright winter sunlight.
As the amount of descriptive text decreases, the AI receives less information when generating an image. The image gradually becomes more distorted with the reduction of text: elements disappear, and the composition becomes simpler and flatter. Even the people depicted within grow increasingly different from those in the original picture.
This phenomenon may reveal the large language model’s heavy reliance on language, where language serves as an absolute variable in the process of image generation.
5. AI Extension
AI Generated
My Photo
By using Adobe Photoshop’s built-in AI image generation feature, we can extend the edges of our photographs and add additional elements. Even upon close inspection, traces of AI generation often remain visible.
However, if the original image features highly regular and repetitive patterns—such as bricks, railings, or leaves—the AI finds it easier to extend them logically, making the generated portions much harder to distinguish from the original.
6. AI Coloring
My Photo
AI Colored
When coloring black-and-white photographs, AI mainly relies on differences in grayscale and the recognition of specific objects—for example, identifying trees and grass as green. Generally, the colors produced by AI have relatively low saturation and tend to be conservative, avoiding overly drastic changes. However, in certain situations, the coloring process can fail. For instance, in the example below, smoke is mistakenly tinted blue-green due to the strong contrast with the bright sky.
AI also struggles to recognize details in blurry photos As a result, it often defaults to applying a uniform brown tone, giving the appearance of coloring without actually enhancing the image meaningfully.
7. Change in Texture
My Photo
AI Generated
Christo and Jeanne-Claude’s works are remarkable. By using AI, I can apply this idea of texture-based wrapping to almost anything—a human figure, a giant Buddha, or even a massive cloud.
There are so many more possibilities for manipulating objects within photographs, creating a wide range of effects. I often feel that my identity as a photographer and filmmaker acts as an anchor in my mind—whenever I encounter a thought about art-making, the final form almost always manifests itself as a photograph or a film, even if the idea could also take the shape of an exhibitable sculpture or a piece of performance art. With AI, many of these transformations can now be achieved with just a few clicks.