{"id":13445368,"url":"https://github.com/hou-yz/MVDet","last_synced_at":"2025-03-20T21:30:26.876Z","repository":{"id":42184429,"uuid":"235071940","full_name":"hou-yz/MVDet","owner":"hou-yz","description":"[ECCV 2020] Codes and MultiviewX dataset for \"Multiview Detection with Feature Perspective Transformation\". ","archived":false,"fork":false,"pushed_at":"2024-05-11T10:39:39.000Z","size":3646,"stargazers_count":161,"open_issues_count":2,"forks_count":29,"subscribers_count":5,"default_branch":"master","last_synced_at":"2024-08-01T05:14:29.496Z","etag":null,"topics":["dataset","detection","detector","multiview","pedestrian-detection","pytorch"],"latest_commit_sha":null,"homepage":"https://hou-yz.github.io/publication/2020-eccv2020-mvdet","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/hou-yz.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2020-01-20T10:08:56.000Z","updated_at":"2024-07-09T08:22:57.000Z","dependencies_parsed_at":"2024-05-11T11:46:37.502Z","dependency_job_id":null,"html_url":"https://github.com/hou-yz/MVDet","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hou-yz%2FMVDet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hou-yz%2FMVDet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hou-yz%2FMVDet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hou-yz%2FMVDet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/hou-yz","download_url":"https://codeload.github.com/hou-yz/MVDet/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":221807671,"owners_count":16883631,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["dataset","detection","detector","multiview","pedestrian-detection","pytorch"],"created_at":"2024-07-31T05:00:31.787Z","updated_at":"2024-10-28T08:30:51.484Z","avatar_url":"https://github.com/hou-yz.png","language":"Python","funding_links":[],"categories":["六、PV2BEV method summary"],"sub_categories":["2. Summary of PV2BEV methods based on Hough transform"],"readme":"# Multiview Detection with Feature Perspective Transformation [[Website](https://hou-yz.github.io/publication/2020-eccv2020-mvdet)] [[arXiv](https://arxiv.org/abs/2007.07247)]\n\n```\n@inproceedings{hou2020multiview,\n  title={Multiview Detection with Feature Perspective Transformation},\n  author={Hou, Yunzhong and Zheng, Liang and Gould, Stephen},\n  booktitle={ECCV},\n  year={2020}\n}\n```\n\nPlease 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!\n\n## Overview\nWe release the PyTorch code for **MVDet**, a state-of-the-art multiview pedestrian detector; and **MultiviewX** dataset, a novel synthetic multiview pedestrian detection datatset.\n\nWildtrack             |  MultiviewX\n:-------------------------:|:-------------------------:\n![alt text](https://hou-yz.github.io/images/eccv2020_mvdet_wildtrack_demo.gif \"Detection results on Wildtrack dataset\")  |  ![alt text](https://hou-yz.github.io/images/eccv2020_mvdet_multiviewx_demo.gif \"Detection results on MultiviewX dataset\")\n\n \n## Content\n- [MultiviewX dataset](#multiviewx-dataset)\n    * [Download MultiviewX](#download-multiviewx)\n    * [Build your own version](#build-your-own-version)\n- [MVDet Code](#mvdet-code)\n    * [Dependencies](#dependencies)\n    * [Data Preparation](#data-preparation)\n    * [Training](#training)\n\n\n\n## MultiviewX dataset\nUsing 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**. \n\n![alt text](https://hou-yz.github.io/images/eccv2020_mvdet_multiviewx_dataset.jpg \"Visualization of MultiviewX dataset\")\n\nMultiviewX 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. \n\n### Download MultiviewX\nPlease refer to this [link](https://1drv.ms/u/s!AtzsQybTubHfhYZ9Ghhahbp20OX9kA?e=Hm9Xdg) for download.\n\n### Build your own version\nPlease refer to this [repo](https://github.com/hou-yz/MultiviewX) for a detailed guide \u0026 toolkits you might need.\n\n\n\n\n## MVDet Code\nThis repo is dedicated to the code for **MVDet**. \n\n![alt text](https://hou-yz.github.io/images/eccv2020_mvdet_architecture.png \"Architecture for MVDet\")\n\n### Dependencies\nThis code uses the following libraries\n- python 3.7+\n- pytorch 1.4+ \u0026 tochvision\n- numpy\n- matplotlib\n- pillow\n- opencv-python\n- kornia\n- matlab \u0026 matlabengine (required for evaluation) (see this [link](/multiview_detector/evaluation/README.md) for detailed guide)\n\n### Data Preparation\nBy 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. \n\nYour `~/Data/` folder should look like this\n```\nData\n├── MultiviewX/\n│   └── ...\n└── Wildtrack/ \n    └── ...\n```\n\n### Training\nIn order to train classifiers, please run the following,\n```shell script\nCUDA_VISIBLE_DEVICES=0,1 python main.py -d wildtrack\n``` \nThis should automatically return evaluation results similar to the reported 88.2\\% MODA on Wildtrack dataset. \n\n### Pre-trained models\nYou can download the checkpoints at this [link](https://1drv.ms/u/s!AtzsQybTubHfhNRE9Iy8IjsGMXB17A?e=CCqhIQ).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhou-yz%2FMVDet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhou-yz%2FMVDet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhou-yz%2FMVDet/lists"}