8000 GitHub - JiangMinyang/JCCL: A smartly simple way for joint crowd counting and localization
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

JiangMinyang/JCCL

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Setup

Requirement: Cuda 10.0
Install dependencies: pip install -r requirements.txt
Install DCNv2: cd models/DCNv2 & ./make.sh

The folder structure should be as follow:

datasets
    |-----Demo
            |------img
                    |-----Demo1.jpg
                    ...
                    |-----Demo99.jpg
            |------label
                    |-----Demo1.txt
                    ...
                    |-----Demo99.txt
            |------dataset_config.py
            |------demo_dataloader.py
            |------loading_data.py
    |-----exp
            |------JHU_CKPT
                    |--------jhu_ckpt.pth
    ...

Run

Demo can be ran using python3 demo.py The localization output will be placed under output/JHU_CKPT directory

About

A smartly simple way for joint crowd counting and localization

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
0