EgoBlur: a set of AI models designed to preserve privacy by detecting and blurring PII from images

Access EgoBlur Gen 1 model and tools below

Access EgoBlur Gen 2 model and tools below

INTRODUCING EGOBLUR GEN 2

EgoBlur Gen 2 protects user privacy by detecting and blurring PII from images on egocentric devices.

EgoBlur Gen 2 is Meta’s next-generation open-source solution designed to preserve privacy in egocentric video and imagery. Building on the success of EgoBlur Gen 1, this update suite provides researchers and developers with the tools to automatically detect and obscure personally identifiable information (PII), such as faces and license plates, captured by Project Aria Gen 2 devices and other head mounted cameras.

How is EgoBlur Gen 2 Evaluated?

To evaluate performance and reduce bias, EgoBlur Gen 2 is benchmarked against the Aria Gen 2 Pilot Dataset and CCV2 Dataset.

Self-reported ‘responsible AI labels’ from the CCV2 dataset are used to evaluate EgoBlur Gen 2 against a number of attributes such as skin tone, self-identified gender, age, and country. This helps to ensure EgoBlur works consistently for everybody.

Key Enhancements with EgoBlur Gen 2

EgoBlur Gen 2 is tuned for egocentric data, and trained using data from Aria Gen 2 hardware. Plus native support for high-resolution VRS, PNG, JPEG, and MP4 files.

Unlike EgoBlur Gen 1, the Gen 2 command-line tool automatically preserves the input video’s original FPS, ensuring the blurred output remains perfectly synced with the source material.

Read the EgoBlur Gen 1 Research Paper

For more information about the EgoBlur Gen 1 model, read our paper on arXiv.

READ THE PAPER

BibTex Citation

If you use the EgoBlur Model in your research, please cite the following:

@misc{raina2023aria,
  title		={EgoBlur Model},
  author	={Nikhil Raina and Guruprasad Somasundaram and Kang Zheng and Sagar Miglani and Steve Saarinen and Jeff Meissner and Mark Schwesinger and Luis Pesqueira and Ishita Prasad and Edward Miller and Prince Gupta and Mingfei Yan and Richard Newcombe and Carl Ren and Omkar Parkhi},
  year		={2023},
  eprint	={2308.13093},
  archivePrefix	={arXiv},
  primaryClass	={cs.CV}
}

Frequently Asked Questions

Can I use the models for commercial purposes?
Yes. Both face and license plate models are licensed under Apache 2.0, meaning the models are available for use for both research and industry applications.

How big are the EgoBlur models?
Both the EgoBlur face and license plate models are approximately 400 MB and have ~104 million parameters.

How long did it take to train the models?
The EgoBlur face model takes approximately 7 days to train on 4 machines with 8 NVIDIA V100 GPUs. The license plate model takes about a day to train on similar configurations.

Can the models be used to identify individuals?
No. Like other open source models, such as RetinaFace, the EgoBlur models are trained only to locate the position of faces and license plates of vehicles within color or greyscale images. The models are not used to track or identify individual faces or license plates.

Do the models work on non-egocentric devices e.g. action cameras and mobile phones?
Yes, in addition to egocentric data from Project Aria, the EgoBlur models are trained on non-egocentric data from the CCV2 dataset. The models should give comparable performance to state-of-the-art models like RetinaFace for such non-egocentric data.

What is the model structure?
EgoBlur is based on the Faster RCNN model with a ResNext backbone. The models are trained using Meta’s publicly available Detectron2 and Detectron2go libraries.

How long does the inference take?
Based on an RGB 2 MP image, both the EgoBlur face detection and license plate models run in approximately 0.5 seconds on a GPU, and 8 seconds on a CPU.
Based on a 0.3 MP greyscale image, both face detection and license plate models run in approximately 0.3 seconds on a GPU, and 1.8 seconds on a CPU.

Do the EgoBlur models produce mask labels?
No, EgoBlur face and license plate models are for detection only. The models output rectangular bounding boxes, not masks or labels.

Does EgoBlur work on videos?
Yes, the EgoBlur model works on images, videos, and VRS files.

How can I contact the team?
Please email projectaria@meta.com to report any bugs, or if you have any further queries about the EgoBlur models.