Grand Bassin Kanwar Dataset: Anonymised Imagery of the Maha Shivaratri Pilgrimage, Mauritius (v1.0)

An anonymised, auto-annotated image dataset of the Maha Shivaratri pilgrimage at Grand Bassin (Ganga Talao), Mauritius, focused on pilgrims and the devotional bamboo-and-flower structures (kanwars) they carry. 2,054 JPEG frames (1280x720) across 21 video segments from 9 cameras, recorded 16-17 February 2023. 98,387 COCO-format bounding boxes in three populated categories (pilgrim: 57,408; vehicle: 34,852; kanwar: 6,127); eight further categories are declared for future manual annotation. All annotations are machine-generated (YOLO26m for pilgrim/vehicle; HSV colour saliency + MobileNetV3 classifier for kanwar) and are not human-verified. All frames are anonymised: 3,538 face regions and 33,391 licence-plate regions were pixelated using YuNet/Haar detection plus annotation-driven blurring. Bounding-box geometry is unchanged. Important: frames within a segment are 0.5 s apart; split train/val/test by segment, never by random frame. See DATASET_CARD.md inside the archive for full documentation, ethical-use conditions, and limitations. A dataset descriptor paper is in preparation.