Montreal-based fashion platform SSENSE has partnered with Element AI to release its dataset of expert-annotated images with the AI research community.
The dataset, which will be hosted by Element AI, consists of hundreds of thousands of annotated studio closeups of fashion items in multiple poses, and each item is paired with fine-grained design descriptions sourced from fashion industry professionals. The images are also tagged by attributes like model identity, season, designer, and country.
Element AI said the companies are releasing SSENSE’s dataset because creative artists and designers need more datasets to develop generative models for augmented reality in the fields of art, design, and fashion.
“Fashion is in the midst of a technological revolution, and machine learning is key to unlocking a new universe of possibilities,” said Element AI CEO Jean Francois Gagné. “New technologies are within reach and are now closer to reality thanks to the comprehensive collection of e-commerce images SSENSE is releasing and Element AI’s ability to layer our advanced computer vision and machine learning abilities over the data. ECCV is the right conference to launch and see what’s possible within the Open Machine Learning community.”
As part of the release of SSENSE’s dataset, SSENSE and Element AI are jointly sponsoring a $2,000 text-to-image generation challenge at the European Conference on Computer Vision taking place in Munich, Germany.
Participants of the challenge will receive access to the SSENSE dataset. Element AI and SSENSE will evaluate participants’ generated images using a pre-trained network hooked up to the challenge API and using the opinion of human judges from the AI, art, and fashion fields. Winners will present their work in the inaugural Computer Vision For Fashion, Art and Design workshop at ECCV ‘18.
Those interested in participating in the challenge can register here by July 31.