Code and data for paper:
Multiview Detection with Cardboard Human Modeling
Jiahao Ma*, Zicheng Duan*, Liang Zheng, Chuong Nguyen
[Paper and Appendices]
We propose a new state-of-the-art method MvCHM for multiview object detection, and some major upgrades on existing multiview pedestrian detectiom benchmarks Wildtrack dataset and MultiviewX dataset, respectively named as Wildtrack+ and MultiviewX+.
- Inference Code and Pretrained Models
- Detailed Instructions
- Datasets
- Training Codes
- Results and More visualizations
This project is tested to run on environment with:
- CUDA > 11
- torchvision 0.11.3
- Windowns 10/11, Ubuntu 20.04
- Create a new conda environment named
mvchm
to run this project:
conda create -n mvchm python=3.9.7
conda activate mvchm
- Make sure your system meet the CUDA requirements and install some core packages:
pip install easydict torch==1.12.1+cu113 torchvision==0.13.1+cu113 tqdm scipy opencv-python
- Clone this repository
cd Your-Project-Folder
gir clone git@github.com:ZichengDuan/MvCHM.git