Skip to content
/ DAFC Public

Distribution-aware Forgetting Compensation for Exemplar-Free Lifelong Person Re-identification

Notifications You must be signed in to change notification settings

LiuShiBen/DAFC

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Distribution-aware Forgetting Compensation for Exemplar-Free Lifelong Person Re-identification

arXiv

Introduction

Our work proposes a novel Distribution-aware Forgetting Compensation (DAFC) model that explores cross-domain shared representation learning and domain-specific distribution integration without relying on old exemplars or knowledge distillation.

Getting Started

Requirements

  • Python 3.7
  • torch 1.10.0
  • For more detailed requirements, run
pip install -r requirements.txt

Dataset preparation

  • Please follow Torchreid_Datasets_Doc to download datasets and unzip them to your data path .
  • Prepare the Seen dataset structure as follow: ./docs/seen dataset structure .md
  • Prepare the Unseen dataset structure as follow: ./docs/Unseen dataset structure .md

Training

Training order-1 for training.

python lifelong_train.py --data-dir=/your seen dataset path  --logs-dir==/save path --orders="order-1"

Training order-2 for training.

python lifelong_train.py --data-dir=/your seen dataset path  --logs-dir==/save path --orders="order-2"

About

Distribution-aware Forgetting Compensation for Exemplar-Free Lifelong Person Re-identification

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages