Clothing simulator on moving bodies using recurrent neural networks
- 0 Collaborators
The idea is the following to select a set of clothing items and adapt them to moving bodies (the images are captured by cameras), so that the observer can select which item of clothing best fits their body and how it will look in motion when be on the move. ...learn more
Project status: Concept
Virtual Reality, Robotics, Artificial Intelligence
Intel Technologies
DevCloud,
oneAPI,
Intel Iris Xe,
Intel GPA,
Intel Iris Xe MAX
Overview / Usage
The project in general focuses on the use of neural networks to provide a mechanism by which a user can select a set of clothing items from a bank of images and manage to simulate, through the use of a video, how the clothing fits together on your body. This project tries to solve the difficulties encountered by a user when acquiring a garment by limiting themselves only to two-dimensional projections (through a mirror) of the three profiles xy, xz and yz that limit the reconstruction of how the person would look in three dimensions. and on the move.
Its application will allow a user more satisfied with the options chosen when selecting clothing.