Deep Insight

Repositories

• insightface

  Face Analysis Project on MXNet

  mxnet face-recognition face-detection face-alignment age-estimation arcface retinaface
  Python MIT 2,711 7,766 609 17 Updated Oct 17, 2020

• insightocr

  MXNet OCR implementation. Including text recognition and detection.

  ocr mxnet text-recognition crnn
  Python MIT 35 94 5 0 Updated Sep 17, 2018

• mxnet-SSH

  Reproduce SSH (Single Stage Headless Face Detector) with MXNet

  mxnet face-detection
  Python MIT 21 90 15 0 Updated Jul 31, 2018

• Deformable-ConvNets

  Forked from msracver/Deformable-ConvNets

  Deformable Convolutional Networks
  Python Apache-2.0 901 4 0 0 Updated Feb 10, 2018

• mxnet

  Forked from apache/incubator-mxnet

  Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
  Python Apache-2.0 6,759 4 0 0 Updated Nov 1, 2017

• stn-attention-tensorflow
  Python 2 5 1 0 Updated Sep 1, 2017

• cnnbilstm-mxnet
  Python 2 4 0 0 Updated Sep 1, 2017

• mx-lsoftmax

  Forked from luoyetx/mx-lsoftmax

  mxnet version of Large-Margin Softmax Loss for Convolutional Neural Networks.
  Python BSD-3-Clause 47 1 0 0 Updated Aug 31, 2017

• transfer-mxnet

  transfer learning written in mxnet

  deep-learning mxnet transfer-learning
  Python 9 55 1 0 Updated Aug 17, 2017

• fm-dnn-tensorflow

  Forked from weiweijiuzaizhe/product-nets
  Python 125 20 0 0 Updated Aug 14, 2017

• densenet-pytorch

  Forked from andreasveit/densenet-pytorch

  A PyTorch Implementation for Densely Connected Convolutional Networks (DenseNets)
  Python BSD-3-Clause 113 0 0 0 Updated Jul 26, 2017

• facenet

  Forked from davidsandberg/facenet

  Face recognition using Tensorflow
  Python MIT 4,465 2 0 0 Updated Jul 14, 2017

• image-retrieval

  Instance-level image retrieval, used in street-to-shop scene. Leveraging large-scale noisy data, and clean it automatically. Learning R-MAC descriptors using a larger network such as ResNet101. Train with a siamese architecture to optimize a triplet loss.
  Python 19 52 1 0 Updated Apr 27, 2017