--- title: "Graffiti Drawing Detection Video Dataset" description: "Graffiti dataset contains 550 Full HD videos of graffiti drawings captured outdoors from a static camera at 30 FPS, across approximately 30–40 locations. With metadata…" url: "https://unidata.pro/datasets/graffiti-drawing-detection-video/" date_modified: "2026-09-16T13:18:08+03:00" language: "en-US" --- Graffiti dataset contains 550 Full HD videos of graffiti drawings captured outdoors from a static camera at 30 FPS, across approximately 30–40 locations. With metadata covering action type, camera height, distance, lighting, time of day, and people count, this graffiti dataset supports computer vision, object detection, and graffiti detection systems for smart city and public safety applications. ## Dataset Structure ### The Numbers Section **Numbered list:** - **Number:** 550 — **Text:** Videos ### Tooltips Section **Tooltip items:** - **Name:** Computer Vision - **Name:** Deep Learning - **Name:** Public Safety - **Name:** Object Detection - **Name:** Action Recognition ### Dataset Information **Table with data:** | Characteristic | Data | | --- | --- | | Description | Videos of graffiti drawing on walls, captured outdoors from a static camera perspective | | Data types | Video | | Tasks | Public Safety, Computer Vision | | Total number of files | 500 | | Number of locations | ~30–40 | | Clothing change | Every 10 videos | | Labeling | Metadata (action type, camera height, camera distance, time of day, lighting source, number of people, device, resolution, fps) | **Media Slider:** - **Video on Slayder:** **Link to the sample:** [Download sample](https://drive.google.com/drive/folders/1S-lFQD59uuSIkTcOa2ar1gF4fzhSo7Fd) ### Technical Specifications **Table with data:** | Characteristic | Data | | --- | --- | | Video extensions | MP4 | | Video Resolutions | 1920 x 1080 | | FPS | 30 | | Camera height | 3 m | | Camera distances | 5–10 m | | Device | iPhone 14 | **Source and data collection methodology:** Source and collection methodology: Data was captured via digital cameras ### Statistics - Charts **Charts with Titles:** - **Shortcode:** [ays_chart id='84'] — **caption above the graph:** Time of day - **Shortcode:** [ays_chart id='85'] — **caption above the graph:** Lighting distribution ### Dataset Use Cases - Slider **Industry Cards:** - **Industry:** Smart City — **Title:** Urban Vandalism Monitoring — **Text:** A graffiti detector can help smart city systems identify unauthorized markings across public infrastructure and flag potential vandalism incidents. Computer vision models can distinguish graffiti from surrounding walls, signs, and other urban objects, supporting automated monitoring and faster responses to visible property damage. - **Industry:** Law Enforcement — **Title:** Incident Identification — **Text:** Graffiti detection models can support law enforcement workflows by identifying suspected vandalism in surveillance footage. Automated recognition helps filter large volumes of visual data, highlight relevant incidents, and provide useful evidence for reviewing graffiti activity, repeated offenses, and locations associated with property damage. - **Industry:** Property Management — **Title:** Maintenance Prioritization — **Text:** Graffiti detection systems can assist property managers in identifying unwanted tags and markings that require attention. Automated detection can help organize inspection workflows, prioritize graffiti removals, and track recurring vandalism across buildings, facilities, transport areas, and other managed properties. - **Industry:** AI & Machine Learning — **Title:** Graffiti Recognition Models — **Text:** The dataset provides training material for developing and evaluating machine learning models focused on identifying graffiti and related visual patterns. Researchers can explore object detection, image recognition, video analysis, and automated surveillance approaches for detecting vandalism in different environmental conditions. ### Fact **FAQs Heading:** FAQs **List of Questions:** - **Question:** Can I request a sample of the dataset before purchasing? — **Answer:** Yes. Unidata provides free dataset samples for trial and testing before purchasing the complete dataset. A sample can help you assess video quality, annotation structure, camera perspective, and suitability for your graffiti detector or computer vision workflow. - **Question:** What types of annotations are provided? — **Answer:** The graffiti database includes metadata for action type, camera height, camera distance, time of day, lighting source, number of people, recording device, resolution, and FPS. These annotations can support model training, scenario analysis, and evaluation of graffiti detection systems. - **Question:** What camera perspective is used? — **Answer:** The videos were recorded from a static camera perspective, with the camera positioned approximately 3 meters high and 5–10 meters from the activity. This setup provides a consistent viewpoint for developing and evaluating graffiti detection and action-recognition models. - **Question:** How many locations are represented in the graffiti database? — **Answer:** The dataset covers approximately 30–40 locations. Multiple locations provide variation in the physical surroundings and can help learning models generalize graffiti detection beyond a single environment. - **Question:** Does the dataset include different people and appearances? — **Answer:** Yes. The collection includes variation in participants and clothing, with clothing changed every 10 videos. - **Question:** Is the graffiti dataset real-world or synthetic? — **Answer:** The graffiti dataset contains real-world video footage captured outdoors using real cameras and people performing graffiti-drawing activities. - **Question:** Can the dataset be used for real-time graffiti detection? — **Answer:** Yes. The video data can be used to develop and evaluate models intended for real-time graffiti detection. - **Question:** How are Unidata datasets licensed? — **Answer:** Unidata datasets follow a dual-licensing model: free samples are provided for trial and testing, while complete datasets are available exclusively through purchase. Licensing terms for the full dataset should be reviewed as part of the purchasing process. - **Question:** How are Unidata datasets stored? — **Answer:** Unidata stores datasets securely on AWS cloud infrastructure for high availability and scalability. Its stated storage and management practices are aligned with ISO 27001 and ISO 27701 standards for information security and privacy management. [Full list of this site's AI-readable pages](https://unidata.pro/llms.txt)