In executing the Data Spaces project's image generation process with Midjourney AI, the focus revolves around capturing and encapsulating Singapore's multicultural essence and vibrant heritage within an extensive archive. This begins with a meticulous selection of iconic cultural hubs and the application of precise image capture parameters to ensure the representation of culturally significant scenes. To enhance visual richness, color.adobe is employed to detect and document diverse color palettes within captured images. The subsequent step involves feeding these identified colors and images into the AI image generator, Midjourney.

To streamline this process, three 150px by 150px tiles are sampled from each photograph , strategically chosen from the middle, middle left, and middle right portions, where the focal points of cultural activities often lie in the photograph. This not only creates an efficient archive but also contributes to smoothening the image generation process. The 240 resultant samples are then inputted into Midjourney AI, producing distinctive textile patterns. The amalgamation of these patterns into a captivating tapestry serves as a poignant and artistic homage, deepening the understanding and appreciation of Singapore's rich, harmonious diversity.

The text prompt selection method for the AI in the project demonstrates a thoughtful and streamlined approach to capturing the essence of each area. By simplifying the criteria to two basic characteristics - the most colorful colors in the 10 images of each area and 2-3 descriptive words for each area - we tried to efficiently distill key features for generating meaningful patterns. This method not only simplifies the selection process but also ensures a degree of consistency across diverse cultural hubs. The incorporation of common elements in all prompts, such as the descriptor "Simple flat pattern," "square vector art tile pattern, front facing," the use of specific colors, contributes to creating a cohesive archive with interconnected design elements.

This decision allows for both variety and unity, striking a balance between uniqueness for each area and a shared visual thread that ties the entire tapestry together. For instance, the example prompt, "Simple flat pattern Singapore, international, concrete, business square vector art tile pattern, front facing, using the colors White, Grey, Dark Grey, Blue, without buildings," provides specific criteria while maintaining a standardized structure. This not only streamlines the AI generation process but also contributes to the overarching goal of creating a captivating tapestry that pays homage to Singapore's cultural mosaic.

The layout of the patterns during the printing of the blanket was meticulously crafted to tell a visual story, guiding the viewer's eyes on a journey across the diverse cultural hubs of Singapore. The decision to arrange the patterns in a way that seamlessly transitions from one area to another, such as from Orchard to Fort Canning, reflects a deliberate effort to create a narrative flow akin to a map. The quilt becomes a dynamic tapestry, not only visually representing the unique characteristics of each cultural hub but also inviting the observer to traverse the city's cultural landscape. This thoughtful layout design not only contributes to the aesthetic appeal of the blanket but also serves as a conceptual map, reinforcing the project's essence by physically guiding the viewer through Singapore's rich and harmonious diversity. The result is not just a blanket but a visual journey, where each patterned segment represents a distinct cultural chapter, intricately woven together to form a cohesive and meaningful whole.

The setup was straightforward, employing two lights facing a white backdrop for a scene where a person rolls out of the quilt on the ground and wraps themselves in the fabric. Despite experimenting with various patterned clothes, we found that the multitude of patterns clashed, creating an overwhelming and messy scene. Opting for a blank background resolved this issue. However, we encountered challenges with our camera equipment, particularly with the malfunctioning video function. Resorting to our iPhone for filming, we faced difficulties during video editing, noting discrepancies in color grading and framing due to varying lighting conditions. This is mainly because we have also opted to shoot extra scenes that were not in the original filming space location.

Collaborating with Jahnavi Gupta, a skilled year 1 fashion student, added an experienced touch to the creation of our quilt. The stitching process commenced by trimming the excess fabric left by the printer, forming a margin, followed by marking the designated stitching area using a ruler. Employing a sewing machine and white polyester thread, we seamlessly stitched together the two fabric pieces. With the fabrics joined, we then cut additional cotton fabric, intended to craft a pocket for the polyester stuffing. After completing the stuffing process, we proceeded to stitch four squares each, culminating in the quilting phase of our creative endeavor.