Dict Guided: CVPR 2021 Dictionary-Guided Scene Text Recognition

VinAIResearch published this scene text recognition collection in 2021 for the CVPR conference. It provides images and dictionary-guided labels for text detection and recognition tasks, specifically targeting the VinText benchmark.

Use Cases

• Training OCR models using dictionary-guided labels
• Benchmarking scene text detection on the VinText subset
• Evaluating text recognition accuracy in natural scene environments

Strengths

• Peer-reviewed CVPR 2021 origin
• Includes specialized dictionary-guided annotations
• Supports both detection and recognition tasks

Limitations

• Specific record counts not provided in metadata
• Potential linguistic bias toward Vietnamese text based on VinText tags

Provenance

Source

VinAIResearch

Collection Method

annotated

Freshness

Last updated July 2024; based on 2021 research publication.