It contains video clips of various identity documents like passports, ID cards, and driving licenses [2].
The Mobile Identity Document Video dataset (MIDV) is a series of datasets created to advance computer vision research in document recognition [2]. The "180" version specifically addresses the challenges of capturing identity documents using mobile devices under various conditions [2]. Key Features of MIDV-180
Note: Always check the specific license (such as Creative Commons) attached to the dataset to ensure your use case—whether personal, academic, or commercial—is permitted. Use Cases for MIDV-180
Detecting whether a document presented to a smartphone camera is a real physical card or a digital reproduction on another screen.
Machine learning datasets can often be locked behind expensive paywalls or restricted corporate licensing. Fortunately, the MIDV series is widely known in the academic community for being publicly accessible [2].
Once you download the dataset, there are several powerful applications you can develop:
Because this is an academic dataset, it is hosted on open data platforms rather than commercial websites. You can generally find and download the dataset for free through the following avenues: 1. GitHub Repositories
