UniDataPro/crowd-counting

Crowd Density Dataset - Different Crowd Sizes

The dataset consists of 647 images of crowds, containing up to 11,000 individuals, annotated with keypoints for precise crowd counting and density estimation. It is designed for crowd counting tasks, particularly in crowded scenes settings, accommodating various sizes and challenges in estimating density. The dataset includes examples of both denser crowds and sparser crowds, enhancing counting accuracy for real-world applications in crowd analysis - Get the data

It provides density maps for density estimation, useful in crowd monitoring and object detection tasks. Data supports deep learning models for better crowd control and crowd management, offering feature-rich data such as feature maps and predicted density outputs.

💵 Buy the Dataset: This is a limited preview of the data. To access the full dataset, please contact us at https://unidata.pro to discuss your requirements and pricing options.

Crowd density types:

0-1000, 1000-2000, 2000-3000, 3000-4000, 4000-5000

Additionally, the dataset can help address challenges in public monitoring, especially for crowd gatherings in public places with varying density levels. It's suitable for studies in human detection and detection-based approaches in contexts like video surveillance and public safety.

Metadata for the dataset

Frequently Asked Questions

What types of real-world scenes are represented in the crowd dataset?

The dataset includes images collected from protests, concerts, and other mass events, providing different visual contexts for crowd analysis. These scenes can contain substantial variation in how people are distributed, making the human crowd dataset relevant to models that need to handle both concentrated and dispersed groups. The source diversity can also help when evaluating whether a crowd-counting model generalizes beyond a single event type.

What annotation format is provided for individual people?

The dataset includes JSON annotation files containing key-point labeling for people in the images. Key points provide a representation of individual locations within crowded scenes and can be used as supervision for point-based crowd counting and keypoint detection approaches.

What applications can benefit from this human crowd dataset?

The dataset can support computer vision systems designed to estimate crowd size, analyze crowd density, and detect people in large gatherings. Potential applications include public-space monitoring, smart-city systems, event management, and analysis of mass gatherings. The key-point annotations are particularly useful for models that estimate people counts from densely populated scenes.

🌐 UniData provides high-quality datasets, content moderation, data collection and annotation for your AI/ML projects