本数据集记录了约1200名行人在某地下通道的步行轨迹，以及对通道内一处商店的视觉注意状态的变化。 本数据已进行脱敏处理，不涉及任何个人隐私信息。 This dataset records the walking trajectories of approximately 1200 pedestrians in a underground corridor, as well as changes in their visual attention state towards a store located within the corridor. The dataset does not include any identifiers or private information.
本数据集记录了约1200名行人在某地下通道的步行轨迹，以及对通道内一处商店的视觉注意状态的变化。
本数据已进行脱敏处理，不涉及任何个人隐私信息。
This dataset records the walking trajectories of approximately 1200 pedestrians in a underground corridor, as well as changes in their visual attention state towards a store located within the corridor.
The dataset does not include any identifiers or private information.
一处地下双向通道被选为实地观察点。通道的一侧有一家便利店，周围没有其他商店、障碍或吸引物。在不妨碍行人的前提下，通过在白天非高峰时段进行的45分钟的视频记录，捕获了约1200名行人在商店前的行动。每15帧（0.5秒）手动标记行人的位置。同时，是否有行人转头看商店也在同一时间手动标记。
An underground bi-directional corridor was selected as the field observation site. One convenience store is located on one side of this corridor with no other stores, obstacles, or attractors around. Without obstructing pedestrians, through 45-minute-long video recordings in off-peak hours during the daytime, the movements of about 1200 pedestrians are captured in front of the store. Pedestrian positions are labeled manually for every 15 frames (0.5 seconds) by estimating the points on the ground directly below the bodies. Also, whether the pedestrian turns their head to the store is labelled manually at the same time.
本数据集一共有6个片段。每个片段包括以下文件：
This dataset has a total of 6 segments. Each segment includes the following files:
Yaml文件记录了每个片段的帧速率、坐标转换信息，以及行人进出商店的ID标签等。详情参见ss0605.yaml中的注释。
The Yaml file records the frame rate, coordinate transformation information, and ID tags of pedestrians entering and exiting the store for each segment. For details, see the comments in ss0605.yaml.
文件夹内，每条行人轨迹均单独记录在一个csv文件里。例如，id1.csv记录了ID为1的行人的位置轨迹。请注意，位置信息需要进行坐标转换方可使用，详情参见对应的yaml文件。
Within the folder, each pedestrian's trajectory is recorded separately in a csv file. For example, id1.csv records the position trajectory of the pedestrian with ID 1. Please note that the position information needs to be converted using the coordinate transformation before it can be used. For details, see the corresponding yaml file.
| id | frame | x | y |
|---|---|---|---|
| 3 | 135.0 | 117.00 | 511.00 |
| 3 | 150.0 | 204.00 | 479.00 |
文件夹内，每个行人开始注视与结束注视商店的时刻（单位：帧），均单独记录在一个csv文件里。如果文件内容为空，则表示该行人并未注视商店。
Within the folder, the moments each pedestrian starts and stops gazing at the store (unit: frame) are each recorded separately in a csv file. If the file content is empty, it means the pedestrian did not gaze at the store.
| id | start | end |
|---|---|---|
| 17 | 600 | 720 |