Skip to content
/ YOLACT Public

An adapted and simplified version of original YOLACT in PyTorch

Notifications You must be signed in to change notification settings

kzhoa/YOLACT

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

28 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

YOLACT

pytorch version

An adapted and simplified version of original YOLACT (https://github.com/dbolya/yolact/).

Paper: YOLACT: Real-time Instance Segmentation

Prepare

Original Yolact authors use pretrained weights of backbone 'resnet-101',which can be download from 'resnet101_reducedfc.pth',

The file 'resnet101_reducedfc.pth' should be put under './weights/' directory.

Train

to train a new model with default save folder './weights/'

python train.py 

or indicate one

python train.py  --save_folder weights/

to train with a weight file

python train.py --resume weights/yolact_base_0_30000.pth

About

An adapted and simplified version of original YOLACT in PyTorch

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages