Adversarial Fashion
2024, AI generated Adversarial pattern designs on bespoke silk dress
CHENG Dong (Demi) in consultation with HU Xiaolin
2024, AI generated Adversarial pattern designs on bespoke silk dress
CHENG Dong (Demi) in consultation with HU Xiaolin
About the Artwork:
Computer scientist Xiaolin Hu and his team have developed a series of AI generated fabric patterns called ‘Adversarial Textures’, specially designed to fool AI Person Detectors used in CCTV systems. These systems are used to capture and analyze images and video footage to automatically detect certain characteristics of people in crowds.
However, the AI system can make mistakes when receiving deliberately designed patterns known as ‘Adversarial Textures’. For this cabinet project, fashion designer Cheng Dong (Demi), who specialises in sustainable fashion, pattern design, and fashion material design, has been commissioned to create a bespoke fashion design using Adversarial textures custom generated by Xiaolin Hu
Computer scientist Xiaolin Hu and his team have developed a series of AI generated fabric patterns called ‘Adversarial Textures’, specially designed to fool AI Person Detectors used in CCTV systems. These systems are used to capture and analyze images and video footage to automatically detect certain characteristics of people in crowds.
However, the AI system can make mistakes when receiving deliberately designed patterns known as ‘Adversarial Textures’. For this cabinet project, fashion designer Cheng Dong (Demi), who specialises in sustainable fashion, pattern design, and fashion material design, has been commissioned to create a bespoke fashion design using Adversarial textures custom generated by Xiaolin Hu
CHENG Dong:
Cheng Dong is a fashion designer whose specialises in sustainable fashion, pattern design, and
fashion material design. Her work have been exhibited at the Shanghai Textile and Clothing
Museum of the National Art Fund, and also showcased at Beijing Fashion Week, Hong Kong
Fashion Week, and Macau Fashion Week. She teaches fashion design teacher at Wuhan
Vocational College of Commerce, and holds an MA in fashion from the school of fashion and
textiles at Hong Kong Polytechnic University. Cheng is currently a PhD candidate in Fine Art at the
University of Macau.
Cheng Dong is a fashion designer whose specialises in sustainable fashion, pattern design, and
fashion material design. Her work have been exhibited at the Shanghai Textile and Clothing
Museum of the National Art Fund, and also showcased at Beijing Fashion Week, Hong Kong
Fashion Week, and Macau Fashion Week. She teaches fashion design teacher at Wuhan
Vocational College of Commerce, and holds an MA in fashion from the school of fashion and
textiles at Hong Kong Polytechnic University. Cheng is currently a PhD candidate in Fine Art at the
University of Macau.
HU Xiaolin:
Hu Xiaolin’s research interests lie at the intersection of deep learning and neuroscience. Inspired by the brain, Hu has proposed novel deep learning models that circumvent many of the difficulties currently faced by current state of the art models. He is also interested in using deep learning techniques to try and unravel the secrets of how the human brain processes complex sensory information. His research lab conducts research on a wide variety of deep learning
applications, such as traffic sign detection, face detection, image segmentation, and medical image analysis, etc. Another recent creative project explores music generation, and aims to automatically generate elements of music including lyrics, melodies and chords.
Hu Xiaolin’s research interests lie at the intersection of deep learning and neuroscience. Inspired by the brain, Hu has proposed novel deep learning models that circumvent many of the difficulties currently faced by current state of the art models. He is also interested in using deep learning techniques to try and unravel the secrets of how the human brain processes complex sensory information. His research lab conducts research on a wide variety of deep learning
applications, such as traffic sign detection, face detection, image segmentation, and medical image analysis, etc. Another recent creative project explores music generation, and aims to automatically generate elements of music including lyrics, melodies and chords.
Explore similar works:
