Skip to content

A multilabel classification method using ensembled vision transformers for single-eye images trained on the ODIR-2019 challenge dataset.

Notifications You must be signed in to change notification settings

madhava20217/Ocular-Disease-Multiclass-Identification-ODIR19-

Repository files navigation

README

Contributors:

  1. Bhagesh Gaur (2020558)
  2. Madhava Krishna (2020217)
  3. Sanyam Goyal (2020116)

The project involves multilabel classification using the ODIR dataset.

The detailed report can be found here: DL Report Final.

The broad steps involved are:

  1. Preprocessing the data to extract single eye-features (credits to Jordi Corbella).
  2. Creating histogram equalised images of the eye-images.
  3. Train Swin-T transformer with early-fusion of image and other features (age, sex).
  4. Using an ensemble of Swin-T vision transformers.

About

A multilabel classification method using ensembled vision transformers for single-eye images trained on the ODIR-2019 challenge dataset.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published