AI-Team-5: Combine Two Paper using Python

Project building architecture combining R-CNN network in Faster R-CNN paper and Facenet

---

Introduction

we propose some application providing face detection of specific person and face blurring of the others. There is two Paper for achieving job of project application.

Requirements

• Previous Code Analytics from other git-repository about FRCNN,MRCNN
• Build unique RCNN Network for mobile computing environment
• python 3.x

Datasets - Imagenet

ImageNet is an image database organized according to the WordNet hierarchy (currently only the nouns), in which each node of the hierarchy is depicted by hundreds and thousands of images. Currently we have an average of over five hundred images per node. We hope ImageNet will become a useful resource for researchers, educators, students and all of you who share our passion for pictures.

Train Image Dataset

• download dataset from this link and put it in this project

Test Image Dataset

The test image dataset are sampled from this link and put ti in this project

TODO

• Reading Papers(R-CNN, Fast R-CNN, Faster R-CNN, FaceNet) and Studying
• Analyze existing code from github "https://github.com/jooyounghun/tensorpack/tree/master/examples/FasterRCNN"
• Build RCNN architecture

DONE

• "FaceNet: A Unified Embedding for Face Recognition and Clustering" Paper
• "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks" Paper
• "Mask R-CNN" Paper

Architecture of Network(Faster R-CNN & Mask R-CNN)

Faster-RCNN

• Key Architecture "Region Proposal Network"

• Detail of RPN

Mask-RCNN

• RoI Align: 2d interpolation for high accuracy of segmentation

• Result view from paper

Reference

• Florian Schroff, Dmitry Kalenichenko, James Philbin. Google Inc. FaceNet: A Unified Embedding for Face Recognition and Clustering. arXiv:1503.03832, 2015
• Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. arXiv:1506.01497, 2016
• Kaiming He Georgia Gkioxari Piotr Dollar Ross Girshick. Facebook AI Research (FAIR). Mask R-CNN. arXiv:1703.06870, 2018