--- title: "Abandoned Object Detection Video Dataset" description: "Abandoned object detection dataset includes 1,062 Full HD videos featuring 50 object types across diverse public scenes. Recorded at 30 FPS with metadata covering camera…" url: "https://unidata.pro/datasets/abandoned-object-detection-video/" date_modified: "2026-09-16T14:09:13+03:00" language: "en-US" --- Abandoned object detection dataset includes 1,062 Full HD videos featuring 50 object types across diverse public scenes. Recorded at 30 FPS with metadata covering camera and scene conditions, it provides video data for developing computer vision models that identify unattended objects such as bags and luggage in surveillance footage. ## Dataset Structure ### The Numbers Section **Numbered list:** - **Number:** 1,062 — **Text:** Videos ### Tooltips Section **Tooltip items:** - **Name:** Computer Vision - **Name:** Deep Learning - **Name:** Public Safety - **Name:** Object Detection - **Name:** Smart Cities ### Dataset Information **Table with data:** | Characteristic | Data | | --- | --- | | Description | Videos of simulated abandoned object scenarios, in which a person leaves an object in the frame and walks away, captured outdoors from a static camera perspective | | Data types | Video | | Tasks | Public Safety, Computer Vision | | Total number of files | 1,062 | | Number of locations | ~60–70 (no more than 3 videos per location) | | Clothing change | Every 10 videos | | Number of objects | 50 | | 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/1dH2zj_OW5qc_azpSaTurBGQkjQQt4bIc) ### Technical Specifications **Table with data:** | Characteristic | Data | | --- | --- | | Video extension | MP4 | | Video resolution | 1920 x 1080 | | FPS | 30 | | Camera height | 3 m | | Camera distances | 5–7 m | | Device | iPhone 14 | **Source and data collection methodology:** Source and collection methodology: Data was captured via digital cameras in controlled, simulated environments ### Statistics - Charts **Charts with Titles:** - **Shortcode:** [ays_chart id="86"] — **caption above the graph:** Time of day - **Shortcode:** [ays_chart id='87'] — **caption above the graph:** Lighting distribution ### Dataset Use Cases - Slider **Industry Cards:** - **Industry:** Public Safety — **Title:** Unattended Object Monitoring — **Text:** Video surveillance systems can monitor public areas for unattended bags, luggage, and other objects that remain in place. Training models with this video data helps develop detection methods for separating stationary objects from normal activity, supporting faster identification of suspicious items in airports, stations, terminals, and other busy environments efficiently. - **Industry:** Transportation — **Title:** Passenger Area Monitoring — **Text:** Airport and railway operators can apply abandoned object detection to passenger areas where unattended luggage may create operational concerns. Computer vision models trained on varied scenes can examine surveillance videos, track objects across frames, and alert staff when a bag or other item appears separated from nearby people at scale. - **Industry:** Lost & Found — **Title:** Unattended Belongings Identification — **Text:** Lost and found services can use object recognition models to locate unattended personal belongings before they become formally reported as missing. Video analysis can help identify bags and other objects, connect observations across surveillance cameras, and provide useful information for staff investigating when and where an item was left behind. - **Industry:** Research & Development — **Title:** Detection Model Evaluation — **Text:** The dataset can support security research focused on identifying suspicious objects in complex public scenes. Researchers can evaluate detection accuracy, compare object detection algorithms, and study challenges involving small objects, occlusion, changing backgrounds, and different viewing conditions when developing video-based recognition systems across varied security monitoring and research applications. ### Fact **FAQs Heading:** FAQs **List of Questions:** - **Question:** Can I request a sample of the ABODA dataset before purchasing or downloading it? — **Answer:** Yes, you can request a sample of the dataset before purchasing the complete dataset. Free samples are available for trial and testing so you can evaluate the video quality, scenes, object types, and annotation structure for your intended application. - **Question:** What types of annotations are provided in the dataset? — **Answer:** The dataset provides metadata annotations covering action type, camera height, camera distance, time of day, lighting source, number of people, recording device, resolution, and FPS. These labels provide contextual information that can be used to analyze detection performance across different surveillance scenes and recording conditions. - **Question:** Is the dataset suitable for training abandoned object detection models? — **Answer:** Yes, the dataset can be used to train and evaluate computer vision models designed to detect abandoned objects in video. Its multiple locations, 50 objects, varying numbers of people, and contextual metadata can support training models for different detection scenarios. - **Question:** Does the dataset contain recordings from multiple locations? — **Answer:** Yes, the dataset includes recordings from approximately 60–70 locations, with no more than three videos per location. - **Question:** What video specifications are provided? — **Answer:** The dataset contains MP4 videos recorded at 1920 × 1080 resolution and 30 FPS. The recordings were captured using an iPhone 14, providing consistent video characteristics for computer vision workflows. - **Question:** How long does it take to receive the dataset? — **Answer:** Once you submit a request, Unidata will reach out to you to review the details and complete the necessary documents. After signing and payment, the dataset will be delivered within 3–10 days. - **Question:** How are Unidata datasets stored? — **Answer:** Unidata stores all datasets securely on AWS cloud infrastructure, ensuring high availability and scalability. Our storage and management practices are aligned with ISO 27001 and ISO 27701 standards, which guarantee compliance with internationally recognized information security and privacy management requirements. - **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. This allows users to evaluate a dataset before obtaining the complete collection. [Full list of this site's AI-readable pages](https://unidata.pro/llms.txt)