8000 GitHub - MUST-AI-Lab/FFPNet: The official codes for "Frequency Feature Pyramid Network with Global-Local Consistency Loss for Crowd and Vehicle Counting in Congested Scenes"
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

The official codes for "Frequency Feature Pyramid Network with Global-Local Consistency Loss for Crowd and Vehicle Counting in Congested Scenes"

Notifications You must be signed in to change notification settings

MUST-AI-Lab/FFPNet

Repository files navigation

FFPNet-pytorch

This is the PyTorch version repo for The official codes for "Frequency Feature Pyramid Network with Global-Local Consistency Loss for Crowd and Vehicle Counting in Congested Scenes".

Datasets

ShanghaiTech part A Dataset.

Prerequisites

We strongly recommend Anaconda as the environment.

Python: 2.7

PyTorch: 0.4.0

CUDA: 9.2

Ground Truth

Please follow the 'gt_A_train.m' and 'gt_A_test.m' to generate the ground truth.

Training Process

Try time ./py_train.h to start training process.

Results

MAE: 57.46 MSE: 88.04

Model

Google Drive

References

If you find the FFPNet useful, please cite our paper. Thank you!

bit citation

About

The official codes for "Frequency Feature Pyramid Network with Global-Local Consistency Loss for Crowd and Vehicle Counting in Congested Scenes"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
0