Developing an AI-based automated fashion design system_compressed.pdf

Department of Information Technology Government Engineering College, Sreekrishnapuram Developing an AI-based automated fashion design system Presented By: Guide: Dr.Rani M R...

Department of Information Technology Government Engineering College, Sreekrishnapuram Developing an AI-based automated fashion design system Presented By: Guide: Dr.Rani M R Andriya Albert (PKD21IT014) 1  CONTENTS INTRODUCTION COMPARISON GENERATIVE AI IN FASHION APPLICATIONS CHALLENGES AND LIMITATIONS RESEARCH OBJECTIVES SYSTEM DEVELOPMENT FUTURE DIRECTIONS CONCLUSION REFERENCES 2 INTRODUCTION AI is transforming many sectors, from healthcare to finance, by automating tasks, analyzing large datasets, and even generating creative content. Focus on Fashion Industry Its impact on the fashion industry is growing day by day. It’s becoming a creative partner in the design process. AI brings, new possibilities such as enhanced trend forecasting, personalized fashion, and more efficient production. It enhance human creativity and speeds up the design process, hence reducing time-to-market. 3 COMPARISON 4 AI in Fashion Design Generative AI in Fashion Generative AI refers to algorithms that can generate new content, such as images, designs, and patterns, based on the data they’ve been trained on. In fashion, this means creating new clothing designs, prints, or even entire collections. Some of the key technologies include GANs (Generative Adversarial Networks) and StyleGAN, which are powerful for generating high-quality, realistic images and designs. 6  Applications Design Creation Customization. Generative AI can create new fashion AI can generate customized designs by learning from vast datasets designs based on individual of existing fashion styles, trends, and preferences, body types, or specific historical designs. customer requirements, offering a high level of personalization. 7 Challenges and Limitations 1 Limited Creativity 2 Repetition Risk AI generates designs based AI might produce designs that on existing patterns and data are too similar to existing ones, but doesn’t create truly lacking originality. original ideas. 3 Quality of Data 4 Bias and Representation AI-generated designs rely on the If AI isn’t trained with a wide variety quality of the training data. Poor of examples, AI might produce data can result in designs that designs that lack cultural or body don’t meet consumer needs. type inclusivity. 8  1. Analyze Existing Tools Research and examine current AI- based garment design tools. Identify differences between AI-driven and human-driven design methods. Research 2. Develop an AI-Based System Objectives Create a system that integrates AI with human fashion knowledge. Ensure the system is user-friendly and adaptable to industry needs. 9 What Are GANs ? (Generative Adversarial Networks) 10  System Development Using StyleGAN2 Pilot Program 11 1. Using StyleGAN2 StyleGAN2 is a powerful AI model that creates realistic and varied images, making it ideal for fashion design. Why Choose StyleGAN2 ? Realistic Designs: It creates images that resemble real clothes, ensuring that the designs are visually appealing. Variety: It generates a wide range of styles and patterns, encouraging creativity and exploration. Control: Designers can adjust specific aspects like colors and textures to align with their vision. BACK TO AGENDA PAGE 12 2. Pilot Program A pilot program was created to test the new design system in a real- world setting with fashion designers. Fashion designers are actively using the system and providing feedback. This helps to identify what works well and what needs improvement. The feedback will be used to refine the system, making it more practical and useful for fashion design. BACK TO AGENDA PAGE 13 BACK TO AGENDA PAGE 14 Human Design Process : Analyzing Data: Humans start by analyzing data relevant to garment design. Determining the Concept: The concept of the garment is determined based on the analyzed data. Creating Designs: The actual design process begins where initial garment designs are created. Modifying the Design: Designs are modified and refined as needed. Finalizing the Design: The design is finalized before moving to production. 15 AI design process: Module 1: Internal Data Collection Collects user preferences, product data, and local fashion knowledge. Module 2: External Data Collection Gathers and analyzes fashion trends from the web, runway shows, and large-scale trend reports. Module 3: Design Source Database Keep a library of keywords and images that might be useful for designing garments. Module 4: Design Generation and Modification Generate and modify garment designs based on the collected data. 16 17  1. Enhancing AI Capabilities: Improve the data used for training AI to make designs more accurate and diverse. Future Make AI tools easier for designers to use and collaborate with in real-time. Directions. 2. Industry Adaptation: Make AI systems scalable for use in different fashion sectors. Customize AI to fit the needs of different brands. 18 CONCLUSION AI is transforming fashion design by boosting both creativity and efficiency. The AI-based system we discussed combines AI capabilities with human creativity, providing practical tools that enhance design processes and decision- making in the fashion industry. As AI technology advances, it will play a crucial role in driving innovation and shaping the future of fashion. 19 REFERENCES Choi, W., Jang, S., Kim, H.Y., Lee, Y., Lee, S.G., Lee, H. and Park, S., 2023. Developing an AI- based automated fashion design system: reflecting the work process of fashion designers. Fashion and Textiles, 10(1), p.39. An, H., Lee, G, Y., Choi, Y., & Park, M. 2023. Conceptual framework of hybrid style in fashion image datasets for machine learning. Fashion and Textiles, 10, 18. https://www.wikipedia.org/ https://chatgpt.com/ 20 THANK YOU 21