If you are looking for specific subsets or similar large-scale identity datasets, you may also find these helpful: : Focuses on low-light and high-distortion scenarios.
: A public dataset used for training AI to detect holographic security features on identity documents. Research Focus
: Benchmarks for document localization, semantic segmentation, and OCR (Optical Character Recognition) under low light or strong projective distortions. Компьютерная оптика Related Datasets & Papers
: The original baseline dataset featuring 50 different document types. ResearchGate pre-trained models that utilize these MIDV datasets for document recognition?
Instead of risking malware or legal action by chasing free, unauthorized copies, consider supporting the official release. Whether you rent, stream, or buy the disc, accessing the “full” version legitimately guarantees the best experience—and respects the creators behind the code.
# Install pwntools (latest) pip3 install --user pwntools
MIDV-2020: A Comprehensive Benchmark Dataset for Identity Document Analysis (and its related extensions). Why this paper is relevant