An open-source tool to detect and blur faces and license plates in images for privacy compliance, using TensorFlow object detection.
understand.ai Anonymizer is an open-source tool that automatically detects and blurs faces and license plates in images to protect privacy and support GDPR compliance. It uses pre-trained TensorFlow models to identify sensitive regions and applies configurable blurring techniques. The tool is designed for batch processing of images, particularly those from automotive sensors.
Developers and companies in automotive, surveillance, or data processing industries needing to anonymize visual data for privacy regulations. It's also suitable for researchers handling image datasets containing personal identifiers.
It provides a ready-to-use, free solution for privacy anonymization with pre-trained models, eliminating the need to build detection systems from scratch. The open-source weights and simple CLI make it accessible for integration into existing pipelines.
ARCHIVED An anonymizer to obfuscate faces and license plates.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Provides ready-to-use TensorFlow weights for faces and license plates trained on in-house automotive datasets, saving development time and effort.
Allows adjustment of Gaussian kernel parameters for blur intensity and smooth transitions, offering flexibility in anonymization quality.
Easy command-line interface supports batch anonymization of images from folders with various format extensions, streamlining workflows.
Specifically designed to help organizations anonymize sensitive visual data for privacy regulations, with examples from automotive use cases.
The repository is read-only with no future updates, meaning potential compatibility issues, lack of bug fixes, and no community support.
Explicitly stated to fail on low-quality, grayscale, or non-standard camera images like fish-eye, restricting its use to specific sensor data.
Written for ease-of-use over speed, missing multiprocessing and batched detection, which can slow down processing on large datasets.