Skip to content
A Dual Refinement Mechanism for Real-World Visual Detection
Python C Cuda Shell C++
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data
layers
model
scripts
utils
.gitattributes
.gitignore
LICENSE
README.md
evaluate.py
evaluate_coco.py
evaluate_trn.py
make.sh
multi_eval.py
test.py
test_image.py
test_video.py
test_video_trn.py
train.py
train_trn.py

README.md

A Dual Refinement Mechanism for Real-World Visual Detection

Inrtroduction

This repository is initially based on ssd.pytorch, and some modules are from RFBNet and deformable-convolution-pytorch, a huge thank to them.

We currently only support PyTorch-0.4.0 and CUDA 8.0. ./make.sh is required for DeformableConv and COCO tools during installation.

Models

Our models are aviliable at GoogleDrive and BaiduYun

Train

cd script
./train.sh
./train_trn.sh

Eval

cd script
./batch_eval.sh
./batch_eval_trn.sh

Paper

arXiv paper describes this project in detail.

Xingyu Chen, Junzhi Yu, Shihan Kong, Zhengxing Wu, and Li Wen, "Towards Real-Time Accurate Object Detection in Both Images and Videos Based on Dual Refinement", arXiv:1807.08638, 2018.

You can’t perform that action at this time.