E2E-MFD

E2E-MFD: Towards End-to-End Synchronous Multimodal Fusion Detection

• This code is for horizontal OD based on Detectron2. The oriented OD code is released at E2E-MFD.

Overview

Getting Started

Create the environment

Step 1: Clone the E2E-MFD repository:

To get started, first clone the E2E-MFD repository and navigate to the project directory:

git clone https://github.com/icey-zhang/E2E-MFD-HOD.git
cd E2E-MFD-HOD

Step 2: Environment Setup:

E2E-MFD recommends setting up a conda environment and installing dependencies via pip. Use the following commands to set up your environment:

Create and activate a new conda environment

conda create -n E2E-MFD python=3.9.16
conda activate E2E-MFD

Prepare the dataset M3FD

you can download the dataset and then run

python tools/get_data.py
python txt2xml.py

Training data and test data are divided in the path

EfficientMFD
├── datasets
│   ├── M3FD
│   │   ├── ImageSets
│   │   │   ├── trainval.txt
│   │   │   ├── test.txt
│   │   ├── Annotations
│   │   │   ├── 00000.xml
│   │   │   ├── 00001.xml
│   │   │   ├── ......
│   │   ├── JPEGImages
│   │   │   ├── 00000.mat
│   │   │   ├── 00001.mat
│   │   │   ├── ......
│   │   ├── M3FD_Fusion
│   │   │   ├── ir
│   │   │   ├── vi
│   │   ├── M3FD_Detection
│   │   │   ├── ir
│   │   │   ├── vi

Begin to train and test

Use the config file with this.

python train_net.py
python test.py

Generate fusion images

python save_test_fusion_V.py

Result

M3FD weights
M3FD logs

If you have any questions, please contact [email protected].