ICPR 2026 Competition on Low-Resolution License Plate Recognition

2026-04-24

AI Summary

This work addresses the challenge of low-resolution license plate recognition (LRLPR) in real-world surveillance scenarios, where factors such as long capture distances, compression artifacts, and poor imaging conditions severely degrade image quality. To bridge the gap between academic research and practical deployment, the authors introduce LRLPR-26, the first dataset tailored for real-world LRLPR applications, comprising multi-frame paired high- and low-resolution license plate images. They also organized the first competition focused exclusively on authentic low-quality data, employing an end-to-end blind evaluation protocol that integrates super-resolution, multi-frame fusion, and character recognition techniques. The challenge attracted 269 participating teams, with 99 submitting valid entries; the top-performing team achieved a recognition accuracy of 82.13%, and four teams surpassed the 80% threshold, substantially advancing the state-of-the-art in both performance and algorithmic robustness for this task.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Large-scale security measurements

Abstract

Low-Resolution License Plate Recognition (LRLPR) remains a challenging problem in real-world surveillance scenarios, where long capture distances, compression artifacts, and adverse imaging conditions can severely degrade license plate legibility. To promote progress in this area, we organized the ICPR 2026 Competition on Low-Resolution License Plate Recognition, the first competition specifically dedicated to LRLPR using real low-quality data collected under operationally relevant conditions. The competition was based on the LRLPR-26 dataset, which comprises 20,000 training tracks and 3,000 test tracks; each training track contains five low-resolution and five high-resolution images of the same license plate. Notably, a total of 269 teams from 41 countries registered for the competition, and 99 teams submitted valid entries in the Blind Test Phase. The winning team achieved a Recognition Rate of 82.13%, and four teams surpassed the 80% mark, highlighting both the high level of competition at the top of the leaderboard and the continued difficulty of the task. In addition to presenting the competition design, evaluation protocol, and main results, this paper summarizes the methods adopted by the top-5 teams and discusses current trends and promising directions for future research on LRLPR. The competition webpage is available at https://icpr26lrlpr.github.io/

Problem

• Low-Resolution License Plate Recognition
• LRLPR
• surveillance
• image degradation
• license plate legibility

Innovation

• Low-Resolution License Plate Recognition
• LRLPR-26 dataset
• paired low/high-resolution images
• real-world surveillance
• ICPR competition