Skip to content

scotwilli/Facedetection-TensorBox

Repository files navigation

Facedetection-TensorBox

Face Detection using deep learning and tensorBox

This development is to detect faces on images using deep learning. It uses TensorBox for training. Output of this algorithm is localization of the faces present in image. The accuracy at 2 lakh iteration is 99.56% on test data and performance is sufficiently good but could be further optimized.

This model is trained using inception-v1 pre-trained weights through train.py of tensorbox. You can find the below link to download the model of 2 lakh iterations and evaluate.py script to test on your images.

https://drive.google.com/drive/folders/0B0GEPp2pSHpnOFl0VHBSTFhWVEE?usp=sharing

The dataset used for training this model can be found from below link

https://exhibits.stanford.edu/data/catalog/sx925dc9385

About

Face Detection using deep learning and tensorBox

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published