Search Results

Search Constraints

Start Over
Filtering by: Creator Rawashdeh, Samir A. Remove constraint Creator: Rawashdeh, Samir A. Type Work Remove constraint Type: Work

Search Results

  • SID: Stereo Image Dataset for Autonomous Driving in Adverse Conditions

    work
    Creator:
    El Shair, Zaid A., Abu-raddaha, Abdalmalek , Cofield, Aaron, Alawneh, Hisham, Aladem, Mohamed, Hamzeh, Yazan, and Rawashdeh, Samir A.
    Description:
    The SID dataset was curated to support advanced research in autonomous driving systems, particularly focusing on perception under adverse weather and lighting conditions. This dataset encompasses over 178k high-resolution stereo image pairs organized into 27 sequences, reflecting a rich variety of conditions such as snow, rain, fog, and low light. It covers dynamic changes in driving scenarios and environmental backgrounds, including university campuses, residential streets, and urban settings. The dataset is designed to challenge perception algorithms with scenarios such as partially obscured camera lenses and varying visibility, promoting the development of robust computer vision models. No specialized software or scripts are necessary for accessing the image data, as the files are provided in standard PNG format. However, researchers and developers may require their image processing and computer vision toolkits to utilize the dataset effectively in their work.
    Keyword:
    Autonomous Driving, Adverse Weather, Stereo Vision, Image Dataset, Computer Vision, and Perception Algorithms
    Citation to related publication:
    El-Shair, Z.A., Abu-raddaha, A., Cofield, A., Alawneh, H., Aladem, M., Hamzeh, Y., and Rawashdeh, S.A., 2024. SID: Stereo Image Dataset for Autonomous Driving in Adverse Conditions. In 2024 IEEE National Aerospace and Electronics Conference (NAECON). IEEE. In press.
    Discipline:
    Engineering

  • MEVDT: Multi-Modal Event-Based Vehicle Detection and Tracking Dataset

    work
    Creator:
    El Shair, Zaid A. and Rawashdeh, Samir A.
    Description:
    The MEVDT dataset was created to fill a critical gap in event-based computer vision research by supplying a high-quality, real-world labeled dataset. Intended to facilitate the development of advanced algorithms for object detection and tracking applications, MEVDT includes multi-modal traffic scene data with synchronized grayscale images and high-temporal-resolution event streams. Additionally, it provides annotations for object detection and tracking with class labels, pixel-precise bounding box coordinates, and unique object identifiers. The dataset is organized into directories containing sequences of images and event streams, comprehensive ground truth labels, fixed-duration event samples, and data indexing sets for training and testing. and To access and utilize the dataset, researchers need specific software or scripts compatible with the data formats included, such as PNG for grayscale images, CSV for event stream data, AEDAT for the encoded fixed-duration event samples, and TXT for annotations. Recommended tools include standard image processing libraries for PNG files and CSV or text parsers for event data. A specialized Python script for reading AEDAT files is available at:  https://github.com/Zelshair/cstr-event-vision/blob/main/scripts/data_processing/read_aedat.py, which streamlines access to the encoded event sample data.
    Keyword:
    Computer Vision, Event-Based Vision, Object Detection, Object Tracking, and Multi-Modal Vision Dataset
    Citation to related publication:
    El Shair, Z. and Rawashdeh, S., 2024. MEVDT: Multi-Modal Event-Based Vehicle Detection and Tracking Dataset. Data In Brief (under review)., El Shair, Z. and Rawashdeh, S.A., 2022. High-temporal-resolution object detection and tracking using images and events. Journal of Imaging, 8(8), p.210., El Shair, Z. and Rawashdeh, S., 2023. High-temporal-resolution event-based vehicle detection and tracking. Optical Engineering, 62(3), pp.031209-031209., and El Shair, Z.A., 2024. Advancing Neuromorphic Event-Based Vision Methods for Robotic Perception Tasks (Doctoral dissertation, University of Michigan-Dearborn).
    Discipline:
    Engineering