Synthetic USA Passport Images for Document Recognition Training

More than 1,000 AI-generated passport images created for training OCR and computer vision systems on identity documents. All records are fully synthetic and do not correspond to real individuals, as stated by the author 'ud-synthetic'. The dataset was last updated on 2026-05-14.

Use Cases

• Train optical character recognition models based on synthetic passport text fields.
• Develop computer vision models for document layout and structure detection based on AI-generated images.
• Benchmark identity document forgery detection algorithms based on synthetic data.
• Test data anonymization and privacy-preserving ML techniques based on the fully synthetic nature of the data.

Strengths

• Contains more than 1,000 AI-generated images, providing a substantial volume for training.
• All data is fully synthetic, mitigating privacy and ethical concerns associated with real identity documents.
• Explicitly intended for training OCR and computer vision systems, indicating a clear purpose.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

ud-synthetic on Hugging Face.

Collection Method

Synthetically generated via AI.

Freshness

Last updated 2026-05-14 15:09:51; freshness should be verified.

Geography

USA (likely based on document format).