Make a machine learning model for virtual try-on using GAN
Budget: $30 – $250 USD
Find or create a machine learning model for virtual try-on for clothing. I have tested many open source virtual try-on projects, and they all have a major limitation: they require the user to be wearing the same type of clothing that they want to try on. For example, if a user wants to try on a tshirt, they must be wearing a tshirt in their selfie photo. That is bad and not what I am looking for. I want the user to be able to try on ANY type of clothing with 1 selfie photo.
One way to accomplish this is to get a SMPL or STAR body model of the user, and then overlay clothing on the SMPL or STAR body model.
Here are the project requirements:
- The user uploads 1 full-body selfie photo wearing underwear or tight clothing.
- You have 1 photo of a piece of clothing.
- The system overlays the clothing on the user's body in a realistic manner.
- The system supports ANY clothing type (e.g. shirts, pants, shorts, sweaters, jackets, dresses, etc.)
One way to accomplish this is to get a SMPL or STAR body model of the user, and then overlay clothing on the SMPL or STAR body model.
Here are the project requirements:
- The user uploads 1 full-body selfie photo wearing underwear or tight clothing.
- You have 1 photo of a piece of clothing.
- The system overlays the clothing on the user's body in a realistic manner.
- The system supports ANY clothing type (e.g. shirts, pants, shorts, sweaters, jackets, dresses, etc.)