Skip to content

CMPE593 Term Project. Replication of the paper titled "FADE: A Task-Agnostic Upsampling Operator for Encoder-Decoder Architectures" for object detection task.

Notifications You must be signed in to change notification settings

kutayeroglu/fade-replicate

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

30 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FADE Replicate

CMPE593 - Deep Learning for Computer Vision, Term Project This project focuses on replicating FADE, a task-agnostic upsampling operator for encoder-decoder architectures.

While FADE is designed for a variety of tasks, this implementation specifically covers object detection.

  • Notebooks under "main-notebooks" were ran in Google Colab
  • "src" folder contains all necessary functions which are called in main notebooks.
  • "src" folder and its contents were placed on the path defined in the notebooks for import.
  • Only file that is not mine is named "FADE_H2L.py" provided by the original authors of the paper.
    • Can be found in source repository listed in Resources section.

Resources

About

CMPE593 Term Project. Replication of the paper titled "FADE: A Task-Agnostic Upsampling Operator for Encoder-Decoder Architectures" for object detection task.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published