Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/hou-yz/MVDet
[ECCV 2020] Codes and MultiviewX dataset for "Multiview Detection with Feature Perspective Transformation".
https://github.com/hou-yz/MVDet
dataset detection detector multiview pedestrian-detection pytorch
Last synced: 12 days ago
JSON representation
[ECCV 2020] Codes and MultiviewX dataset for "Multiview Detection with Feature Perspective Transformation".
- Host: GitHub
- URL: https://github.com/hou-yz/MVDet
- Owner: hou-yz
- Created: 2020-01-20T10:08:56.000Z (almost 5 years ago)
- Default Branch: master
- Last Pushed: 2024-05-11T10:39:39.000Z (6 months ago)
- Last Synced: 2024-08-01T05:14:29.496Z (3 months ago)
- Topics: dataset, detection, detector, multiview, pedestrian-detection, pytorch
- Language: Python
- Homepage: https://hou-yz.github.io/publication/2020-eccv2020-mvdet
- Size: 3.48 MB
- Stars: 161
- Watchers: 5
- Forks: 29
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- Awesome-BEV-Perception - project
README
# Multiview Detection with Feature Perspective Transformation [[Website](https://hou-yz.github.io/publication/2020-eccv2020-mvdet)] [[arXiv](https://arxiv.org/abs/2007.07247)]
```
@inproceedings{hou2020multiview,
title={Multiview Detection with Feature Perspective Transformation},
author={Hou, Yunzhong and Zheng, Liang and Gould, Stephen},
booktitle={ECCV},
year={2020}
}
```
Please visit [link](https://github.com/hou-yz/MVDeTr) for our new work MVDeTr, a transformer-powered multiview detector that achieves new state-of-the-art!
## Overview
We release the PyTorch code for **MVDet**, a state-of-the-art multiview pedestrian detector; and **MultiviewX** dataset, a novel synthetic multiview pedestrian detection datatset.
Wildtrack | MultiviewX
:-------------------------:|:-------------------------:
 | 
## Content
- [MultiviewX dataset](#multiviewx-dataset)
* [Download MultiviewX](#download-multiviewx)
* [Build your own version](#build-your-own-version)
- [MVDet Code](#mvdet-code)
* [Dependencies](#dependencies)
* [Data Preparation](#data-preparation)
* [Training](#training)
## MultiviewX dataset
Using pedestrian models from [PersonX](https://github.com/sxzrt/Dissecting-Person-Re-ID-from-the-Viewpoint-of-Viewpoint), in Unity, we build a novel synthetic dataset **MultiviewX**.
MultiviewX dataset covers a square of 16 meters by 25 meters. We quantize the ground plane into a 640x1000 grid. There are 6 cameras with overlapping field-of-view in MultiviewX dataset, each of which outputs a 1080x1920 resolution image. We also generate annotations for 400 frames in MultiviewX at 2 fps (same as Wildtrack). On average, 4.41 cameras are covering the same location.
### Download MultiviewX
Please refer to this [link](https://1drv.ms/u/s!AtzsQybTubHfhYZ9Ghhahbp20OX9kA?e=Hm9Xdg) for download.
### Build your own version
Please refer to this [repo](https://github.com/hou-yz/MultiviewX) for a detailed guide & toolkits you might need.
## MVDet Code
This repo is dedicated to the code for **MVDet**.
### Dependencies
This code uses the following libraries
- python 3.7+
- pytorch 1.4+ & tochvision
- numpy
- matplotlib
- pillow
- opencv-python
- kornia
- matlab & matlabengine (required for evaluation) (see this [link](/multiview_detector/evaluation/README.md) for detailed guide)
### Data Preparation
By default, all datasets are in `~/Data/`. We use [MultiviewX](#multiviewx-dataset) and [Wildtrack](https://www.epfl.ch/labs/cvlab/data/data-wildtrack/) in this project.
Your `~/Data/` folder should look like this
```
Data
├── MultiviewX/
│ └── ...
└── Wildtrack/
└── ...
```
### Training
In order to train classifiers, please run the following,
```shell script
CUDA_VISIBLE_DEVICES=0,1 python main.py -d wildtrack
```
This should automatically return evaluation results similar to the reported 88.2\% MODA on Wildtrack dataset.
### Pre-trained models
You can download the checkpoints at this [link](https://1drv.ms/u/s!AtzsQybTubHfhNRE9Iy8IjsGMXB17A?e=CCqhIQ).