MIDV-180 is a cornerstone for training several types of machine learning models:
MIDV-180 stands for , featuring 180 video clips. It is an expansion of earlier datasets (like MIDV-500) specifically designed to capture the challenges of mobile document recognition. The dataset focuses on the variability of backgrounds, lighting, and physical document positioning that occurs when a user captures an ID card or passport using a smartphone. The dataset includes: 180 video clips of various identity documents.
The MIDV-180 dataset is widely used in academic research and by developers of OCR (Optical Character Recognition) software. It is typically structured into two main parts:
MIDV-180 is a cornerstone for training several types of machine learning models:
MIDV-180 stands for , featuring 180 video clips. It is an expansion of earlier datasets (like MIDV-500) specifically designed to capture the challenges of mobile document recognition. The dataset focuses on the variability of backgrounds, lighting, and physical document positioning that occurs when a user captures an ID card or passport using a smartphone. The dataset includes: 180 video clips of various identity documents. midv-180
The MIDV-180 dataset is widely used in academic research and by developers of OCR (Optical Character Recognition) software. It is typically structured into two main parts: MIDV-180 is a cornerstone for training several types