##Face++团队
作者:何宜晖
链接:https://www.zhihu.com/question/26558251/answer/113530447
来源:知乎
著作权归作者所有。商业转载请联系作者获得授权,非商业转载请注明出处。
###Recognition
- Channel Pruning for Accelerating Very Deep Neural Networks[1707.06168]
- Learning Deep Face Representation (http://arxiv.org/pdf/1403.2802.pdf)
- Naive-Deep Face Recognition: Touching the Limit of LFW Benchmark or Not? (http://arxiv.org/pdf/1501.04690.pdf)
###Scene Text Understanding - Incidental Scene Text Understanding:Recent Progresses on ICDAR 2015 Robust ReadingCompetition Challenge 4 (http://arxiv.org/pdf/1511.09207.pdf)
- Scene Text Detection via Holistic, Multi-ChannelPrediction (http://arxiv.org/pdf/1606.09002.pdf)
###Face Alignment - Coarse-to-fine Face Alignment with Multi-Scale Local Patch Regression (http://arxiv.org/pdf/1511.04901.pdf)
- Extensive Facial Landmark Localization with Coarse-to-fine Convolutional Network Cascade (http://www.cv-foundation.org//openaccess/content_iccv_workshops_2013/W11/papers/Zhou_Extensive_Facial_Landmark_2013_ICCV_paper.pdf)
###Compression - Learning Compact Face Representation: Packing a Face into an int32 (PDF)
###Recover - Learning Face Hallucination in the Wild (PDF)
###Not related to Face - Exploiting Local Structures with the Kronecker Layer inConvolutional Networks (http://arxiv.org/pdf/1512.09194.pdf)
- DOREFA-NET: TRAINING LOW BITWIDTH CONVOLUTIONALNEURAL NETWORKS WITH LOW BITWIDTHGRADIENTS (http://arxiv.org/pdf/1606.06160.pdf)
- UnitBox: An Advanced Object Detection Network (http://arxiv.org/abs/1608.01471)