One of the Toolkit’s output options is to save the data locally in a CSV file. This is very useful when operating in a fully offline environment, or as a local backup of your data in case of network connection issues.

To use this you just need to enable the preferred type of output and specify a CSV path in the Toolkit’s Settings File.

Each CSV file name will contain the data type (raw, aggregated etc.), Toolkit instance name, and date. Every day at midnight a new CSV file is generated with the same data type and instance name, but a new date.

Example CSV output - aggregated

Text version:

aggregation_time, aggregation_freq, id, initial_detection_time, age, gender, dweller, viewer, viewing_sessions, average_viewing_duration, viewing_duration, dwell_sessions, average_dwell_duration, dwell_duration, detection_duration, smile_viewing, smile_maximum, face_mask

2023-01-24 14:36:35+01:00, 3000, 6, 2023-01-24 14:36:35+01:00, 39, m, 1, 1, 3, 195, 585, 1, 1532, 1532, 1532, 36, 43, 0

Example CSV output - raw

Text version:

frame_number, timestamp, id, age, gender, pitch, yaw, roll, is_viewing, is_dwelling, gaze_x, gaze_y, detection_duration, total_viewing_duration, current_viewing_duration, total_dwell_duration, current_dwell_duration, smile, bounding_box_x, bounding_box_y, bounding_box_width, bounding_box_height, world_x, world_y, world_z, detection_confidence, distance_from_camera, face_mask

1, 2023-01-24 14:36:05.000+01:00, 1, 17, -86, 5, -29, 0, 0, 1, 66, 12, 0, 0, 0, 0, 0, 76, 1062, 216, 94, 98, 74, 15, 114, 0.999081, 138, 0

Example CSV output - POP Aggregated audience per content