Viet Handwriting OCR V2: 60,248 Vietnamese Handwritten Text Images

Vietnamese handwritten text images collected from public internet sources. The dataset contains 60,248 images curated for Handwritten Text Recognition research and applications. It was created by 5CD-AI and last updated on Hugging Face in May 2026.

Use Cases

• Train Vietnamese handwritten text recognition models based on the collection of 60,000+ images.
• Benchmark OCR algorithm performance on Vietnamese script based on the curated dataset.
• Develop applications for digitizing Vietnamese handwritten documents based on the provided image corpus.
• Fine-tune pre-trained vision-language models for Vietnamese text extraction based on the handwritten samples.

Strengths

• Contains 60,248 individual handwritten text images.
• Specifically curated for Vietnamese language OCR, addressing a noted need for local data.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is known, but the distribution of text content, writer demographics, and image quality are unspecified.

Provenance

Source

5CD-AI via Hugging Face.

Collection Method

Images were crawled from public internet sources.

Freshness

Last updated 2026-05-10 23:25:43; freshness should be verified.

Geography

Vietnam 🇻🇳