ViV-ReID: Bidirectional Structural-Aware Spatial-Temporal Graph Networks on Large-Scale Video-Based vessel Re-Identification Dataset

Mingxin Zhang1, Fuxiang Feng1, Xing Fang1, Lin Zhang1, Youmei Zhang1,2,Xiaolei Li1, Wei Zhang1
1 School of Control Science and Engineering, Shandong University 2 School of Mathematics and Statistics, Qilu University of Technology (Shandong Academy of Sciences)

Abstract

Vessel re-identification (ReID) serves as a foundational task for intelligent maritime transportation systems. To enhance maritime surveillance capabilities, this study investigates video-based vessel ReID, a critical yet underexplored task in intelligent transportation systems. The lack of relevant datasets has limited the progress of Video-based vessel ReID research work. We established ViV-ReID, the first publicly available large-scale video-based vessel ReID dataset, comprising 480 vessel identities captured from 20 cross-port camera views (7,165 tracklets and 1.14 million frames), establishing a benchmark for advancing vessel ReID from image to video processing. Videos offer significantly richer information than single-frame images. The dynamic nature of video often leads to fragmented spatio-temporal features causing disrupted contextual understanding, and to address this problem, we further propose a Bidirectional Structural-Aware Spatial-Temporal Graph Network (Bi-SSTN) that explicitly aligns spatio-temporal features using vessel structural priors. Extensive experiments on the ViV-ReID dataset demonstrate that image-based ReID methods often show suboptimal performance when applied to video data. Meanwhile, it is crucial to validate the effectiveness of spatio-temporal information and establish performance benchmarks for different methods. The Bidirectional Structural-Aware Spatial-Temporal Graph Network (Bi-SSTN) significantly outperforms state-of-the-art methods on ViV-ReID, confirming its efficacy in modeling vessel-specific spatio-temporal patterns.

Overview of the ReID dataset

The dataset contains 480 vessel identities captured from 20 camera views across multiple ports.




ViV-ReID Datasets

Video link:https://www.youtube.com/watch?v=pEnqXxjHOI8




ViV-ReID Dataset Download

Dataset Access Procedure:

  1. Step 1: Download Agreement - Please download the data usage agreement: AgreementForViVReID.docx
  2. Step 2: Complete Documentation - Fill in all required fields with accurate institutional details and provide appropriate signatures
  3. Step 3: Submit Agreement - Email the fully executed agreement to our research team at: info@vsislab.com

Dataset Usage Policy:

The ViV-ReID dataset is shared with the research community for non-commercial academic purposes only. We kindly ask that users:

  • Use the data solely for research and educational activities
  • Appropriately cite our work in any resulting publications
  • Respect the terms outlined in the license agreement

We appreciate your cooperation in following these procedures, which help us maintain the integrity of our research distribution while supporting the academic community. Dataset access will be granted upon receipt and review of your completed agreement.