Forensic License Plate Recognition with Compression-Informed Transformers
This work addresses a domain-specific challenge in legal investigations by enhancing license plate recognition under strong compression, representing an incremental advance with practical applications.
The paper tackled the problem of forensic license plate recognition from highly compressed and low-resolution footage, proposing a side-informed Transformer architecture that embeds compression knowledge to improve recognition, achieving up to 8.9 percentage points improvement for severely degraded images.
Forensic license plate recognition (FLPR) remains an open challenge in legal contexts such as criminal investigations, where unreadable license plates (LPs) need to be deciphered from highly compressed and/or low resolution footage, e.g., from surveillance cameras. In this work, we propose a side-informed Transformer architecture that embeds knowledge on the input compression level to improve recognition under strong compression. We show the effectiveness of Transformers for license plate recognition (LPR) on a low-quality real-world dataset. We also provide a synthetic dataset that includes strongly degraded, illegible LP images and analyze the impact of knowledge embedding on it. The network outperforms existing FLPR methods and standard state-of-the art image recognition models while requiring less parameters. For the severest degraded images, we can improve recognition by up to 8.9 percent points.