Skip to content

🔎 Building upon Multi-view Feature Fusion Network (MFFNs), introducing new view perspectives of camouflage objects using novel "color attention."

License

Notifications You must be signed in to change notification settings

nathanjzhao/ColorView

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Exploring Color Modification in Multi-view Feature Fusion Networks

Custom ColorView Figure

Utilizing PyTorch, we build upon Multi-view Feature Fusion Network for Camouflaged Object Detection, taking the existing distance and angle-based viewing perspectives and adding a "color view," applying the model to the ACD1K dataset. As a result, we implement a color attention in order to preserve the model's understanding of color-based features during feature-combination within the model's characteristic Co-attention of Multi-view (CAMV) model.

Full results of this CS 231N final project, along with model improvements resulting from dataset expansion and pixel-based noramlization, can be viewed in the following paper)

About

🔎 Building upon Multi-view Feature Fusion Network (MFFNs), introducing new view perspectives of camouflage objects using novel "color attention."

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published