Dragon Lake Parking Dataset

# Dragon Lake Parking Dataset [https://doi.org/10.5061/dryad.tht76hf5b](https://doi.org/10.5061/dryad.tht76hf5b) The [Dragon Lake Parking (DLP) Dataset](https://sites.google.com/berkeley.edu/dlp-dataset) contains annotated video and data of vehicles, cyclists, and pedestrians inside a parking lot. We collected it by flying a drone above a huge parking lot. Abundant vehicle parking maneuvers and interactions are recorded. To the best of our knowledge, this is the first and largest public dataset designated for the parking scenario (up to April 2022), featuring high data accuracy and a rich variety of realistic human driving behavior. ### Statistics #### Raw video * Length: 3.5 hours * Resolution: 4K * Frame rate: 25 fps #### Parking Area * Size: 140 m x 80 m * Number of spots: \~400 #### Agent Types and Count * Vehicles (normal sedan, medium vehicle like SUV, bus): 1216 * Pedestrians: 3904 * Bicycles: 28 * Motorcycles: 5 ## Description of the data and file structure The raw videos are annotated and converted to JSON format. The dataset has a graph structure with the following components * Agent: An agent is an object that has moved in this scene. It contains the object's dimension, type, and trajectory as a list of instances. * Instance: An instance is the state of an agent at a time step, which includes position, orientation, velocity, and acceleration. It also points to the preceding / subsequent instance along the agent's trajectory. * Frame: A frame is a discrete sample from the recording. It contains a list of visible instances at this time step, and points to the preceding / subsequent frame. * Obstacle: Obstacles are vehicles that never move in this recording. * Scene: A scene represents a consecutive video recording with certain length. It points to all frames, agents, and obstacles in this recording. The entire DLP dataset contains 30 scenes, 317,873 frames, 5,188 agents, and 15,383,737 instances. ## Sharing/Access information Two types of data are available: 1. JSONs will provide you all Instances, Agents, Frames, Obstacles, and Scene data so that you can use our Python toolkit. All coordinates are transformed from UTM to the local coordinates of the parking lot. You can download the JSONs directly from Dryad if your research is about trajectory analysis and/or semantic visualization is enough for your computer vision module. 2. Raw video and ground truth annotation. You ONLY need this if you are working on object detection, tracking, semantic segmentation, or end-to-end models with raw bird's eye view camera data. The annotated trajectories are in UTM coordinates. Please understand that we cannot offer software tools for parsing these data. Since the video data is huge, you have to [submit a request](https://forms.gle/Fw5EKy2cKeYCBssF9) if you need raw videos for your research. ## Code/Software We are releasing a Python toolkit, which provides convenient APIs to query and visualize data. [MPC-Berkeley/dlp-dataset: Dragon Lake Parking Dataset by MPC Lab (github.com)](https://github.com/MPC-Berkeley/dlp-dataset)

# 龙湖停车场数据集（Dragon Lake Parking Dataset）[https://doi.org/10.5061/dryad.tht76hf5b]
[龙湖停车场（Dragon Lake Parking, DLP）数据集](https://sites.google.com/berkeley.edu/dlp-dataset)包含了停车场内车辆、骑行者与行人的标注视频及相关数据。本数据集通过在大型停车场上空搭载无人机航拍采集所得，记录了丰富多样的车辆泊车动作与交互场景。据我们所知，截至2022年4月，本数据集是首个且规模最大的针对泊车场景的公开数据集，具备高精度数据特征与多样化的真实人类驾驶行为表现。
### 统计信息
#### 原始视频
* 时长：3.5小时
* 分辨率：4K
* 帧率：25 fps（帧每秒）
#### 停车场区域
* 尺寸：140米 × 80米
* 车位数量：约400个
#### 目标主体类型与数量
* 车辆（普通轿车、SUV等中型车辆、巴士）：1216个
* 行人：3904个
* 自行车：28辆
* 摩托车：5辆
### 数据与文件结构说明
原始视频已完成标注并转换为JSON格式。本数据集采用图结构，包含以下组成部分：
* 主体（Agent）：指场景中发生过移动的对象，包含物体的尺寸、类型以及以实例列表形式存储的轨迹信息。
* 实例（Instance）：指某一时间步下主体的状态，包含位置、朝向、速度与加速度信息，同时会指向该主体轨迹上的前序与后序实例。
* 帧（Frame）：指航拍记录的离散采样帧，包含当前时间步下所有可见的实例列表，并会指向前序与后序帧。
* 障碍物（Obstacle）：指本次记录中全程未发生移动的车辆。
* 场景（Scene）：指一段具有固定时长的连续航拍记录，包含该记录内的所有帧、主体与障碍物信息。
整个DLP数据集共包含30个场景、317,873帧、5,188个主体以及15,383,737个实例。
### 共享与获取说明
本数据集提供两种类型的数据：
1. JSON格式数据：包含所有实例、主体、帧、障碍物与场景数据，可搭配我们提供的Python工具包使用。所有坐标已从通用横轴墨卡托坐标系（UTM）转换至停车场局部坐标系。若您的研究方向为轨迹分析，或仅需语义可视化用于计算机视觉模块，可直接从Dryad平台下载JSON数据。
2. 原始视频与真值标注：仅当您需要开展目标检测、跟踪、语义分割或基于原始鸟瞰视角相机数据的端到端模型研究时，才需要获取此类数据。标注后的轨迹采用通用横轴墨卡托坐标系（UTM）存储。请注意，我们无法提供用于解析此类数据的软件工具。由于视频数据体量庞大，若您的研究需要原始视频，请[提交申请](https://forms.gle/Fw5EKy2cKeYCBssF9)。
### 代码与软件
我们发布了一款Python工具包，提供便捷的查询与可视化API接口。[MPC-Berkeley/dlp-dataset：MPC实验室出品的龙湖停车场数据集（github.com）](https://github.com/MPC-Berkeley/dlp-dataset)

搜集汇总

背景与挑战

背景概述

Dragon Lake Parking Dataset是一个专注于停车场场景的公共数据集，由加州大学伯克利分校的研究人员通过无人机采集，包含3.5小时4K分辨率的视频数据，总计8.02 GB，涵盖30个场景、超过31万帧和5000多个代理的注释信息。该数据集以JSON格式提供，并附有Python工具包，适用于轨迹预测、行为分析和自动驾驶研究，具有高数据准确性和丰富的真实驾驶行为记录，是截至2022年4月该领域最大且首个公开的数据集。