Dataset of Large Gathering Images for Person Identification and Tracking

This paper presents a large gathering dataset of images extracted from publicly filmed videos by 24 cameras installed on the premises Masjid Al-Nabvi, Madinah, Saudi Arabia. consists raw and processed reflecting highly challenging unconstraint environment. The methodology for building four core phases; that include acquisition videos, extraction frames, localization face regions, cropping resizing detected regions. in consist total 4613 frames obtained video sequences. regions 250 persons data to ensure authenticity presented data. further 8 corresponding each subjects (persons) 2000 images. It portrays unconstrained environment with human faces varying sizes pixel quality (resolution). Since sequences are severely degraded due various unavoidable factors, it can be used as benchmark test evaluate detection recognition algorithms research purposes. We have also gathered displayed records presence who appear frames; temporal context. tracking, finding persons, activity monitoring, crowd counting scenarios.