{"id":13442988,"url":"https://github.com/Mrmoore98/VectorMapNet_code","last_synced_at":"2025-03-20T15:31:42.451Z","repository":{"id":37686218,"uuid":"504777705","full_name":"Mrmoore98/VectorMapNet_code","owner":"Mrmoore98","description":"This is the official code base of VectorMapNet (ICML 2023)","archived":false,"fork":false,"pushed_at":"2023-10-07T13:02:49.000Z","size":155,"stargazers_count":405,"open_issues_count":32,"forks_count":55,"subscribers_count":29,"default_branch":"mian","last_synced_at":"2024-10-28T06:00:21.368Z","etag":null,"topics":["autonomous","deep-learning","map-learning"],"latest_commit_sha":null,"homepage":"https://tsinghua-mars-lab.github.io/vectormapnet/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Mrmoore98.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2022-06-18T07:40:14.000Z","updated_at":"2024-10-25T07:09:00.000Z","dependencies_parsed_at":"2024-01-14T16:16:16.250Z","dependency_job_id":"5409b8c4-3dd8-4ee2-b640-2357fd694ee7","html_url":"https://github.com/Mrmoore98/VectorMapNet_code","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/Mrmoore98%2FVectorMapNet_code","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Mrmoore98%2FVectorMapNet_code/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Mrmoore98%2FVectorMapNet_code/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Mrmoore98%2FVectorMapNet_code/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Mrmoore98","download_url":"https://codeload.github.com/Mrmoore98/VectorMapNet_code/tar.gz/refs/heads/mian","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244640079,"owners_count":20485979,"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":["autonomous","deep-learning","map-learning"],"created_at":"2024-07-31T03:01:54.433Z","updated_at":"2025-03-20T15:31:42.040Z","avatar_url":"https://github.com/Mrmoore98.png","language":"Python","funding_links":[],"categories":["Python","Online HD map construction"],"sub_categories":["2022"],"readme":"# VectorMapNet_code\n**VectorMapNet: End-to-end Vectorized HD Map Learning** ICML 2023\n\nThis is the official codebase of VectorMapNet\n\n\n[Yicheng Liu](https://scholar.google.com/citations?user=vRmsgQUAAAAJ\u0026hl=zh-CN), Yuantian Yuan, [Yue Wang](https://people.csail.mit.edu/yuewang/), [Yilun Wang](https://scholar.google.com.hk/citations?user=nUyTDosAAAAJ\u0026hl=en/), [Hang Zhao](http://people.csail.mit.edu/hangzhao/)\n\n\n**[[Paper](https://arxiv.org/pdf/2206.08920.pdf)] [[Project Page](https://tsinghua-mars-lab.github.io/vectormapnet/)]**\n\n**Abstract:**\nAutonomous driving systems require High-Definition (HD) semantic maps to navigate around urban roads. Existing solutions approach the semantic mapping problem by offline manual annotation, which suffers from serious scalability issues.  Recent learning-based methods produce dense rasterized segmentation predictions to construct maps. However, these predictions do not include instance information of individual map elements and require heuristic post-processing to obtain vectorized maps. To tackle these challenges, we introduce an end-to-end vectorized HD map learning pipeline, termed VectorMapNet. VectorMapNet takes onboard sensor observations and predicts a sparse set of polylines in the bird's-eye view. This pipeline can explicitly model the spatial relation between map elements and generate vectorized maps that are friendly to downstream autonomous driving tasks. Extensive experiments show that VectorMapNet achieve strong map learning performance on both nuScenes and Argoverse2 dataset, surpassing previous state-of-the-art methods by 14.2 mAP and 14.6mAP. Qualitatively, VectorMapNet is capable of generating comprehensive maps and capturing fine-grained details of road geometry. To the best of our knowledge, VectorMapNet is the first work designed towards end-to-end vectorized map learning from onboard observations. \n\n**Questions/Requests:** \nPlease file an [issue](https://github.com/Tsinghua-MARS-Lab/vecmapnet/issues) or send an email to [Yicheng](moooooore66@gmail.com).\n\n\n## Bibtex\nIf you found this paper or codebase useful, please cite our paper:\n```\n@inproceedings{liu2022vectormapnet,\n        title={VectorMapNet: End-to-end Vectorized HD Map Learning},\n        author={Liu, Yicheng and Yuantian, Yuan and Wang, Yue and Wang, Yilun and Zhao, Hang},\n        booktitle={International conference on machine learning},\n        year={2023},\n        organization={PMLR}\n    }\n```\n\n\n# Run VectorMapNet\n\n## Note\n\n\n## 0. Environment\n\nSet up environment by following this [script](env.md)\n\n## 1. Prepare your dataset\n\nStore your data with following structure:\n\n```\n    root\n        |--datasets\n            |--nuScenes\n            |--Argoverse2(optional)\n\n```\n\n### 1.1 Generate annotation files\n\n#### Preprocess nuScenes\n\n```\npython tools/data_converter/nuscenes_converter.py --data-root your/dataset/nuScenes/\n```\n\n## 2. Evaluate VectorMapNet\n\n### Download Checkpoint\n| Method       | Modality    | Config | Checkpoint |\n|--------------|-------------|--------|------------|\n| VectorMapNet | Camera only | [config](configs/vectormapnet.py) | [model link](https://drive.google.com/file/d/1ccrlZ2HrFfpBB27kC9DkwCYWlTUpgmin/view?usp=sharing)      |\n\n\n### Train VectorMapNet\n\nIn single GPU\n```\npython tools/train.py configs/vectormapnet.py\n```\n\nFor multi GPUs\n```\nbash tools/dist_train.sh configs/vectormapnet.py $num_gpu\n```\n\n\n### Do Evaluation\n\nIn single GPU\n```\npython tools/test.py configs/vectormapnet.py /path/to/ckpt --eval name\n```\n\nFor multi GPUs\n```\nbash tools/dist_test.sh configs/vectormapnet.py /path/to/ckpt $num_gpu --eval name\n```\n\n\n### Expected Results\n\n| $AP_{ped}$   | $AP_{divider}$ | $AP_{boundary}$ | mAP   |\n|--------------|----------------|-----------------|-------|\n| 39.8 | 47.7    | 38.8          | 42.1 |\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FMrmoore98%2FVectorMapNet_code","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FMrmoore98%2FVectorMapNet_code","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FMrmoore98%2FVectorMapNet_code/lists"}