SSHA
SSHA copied to clipboard
ElegantGod
face detection and alignment
env： gtx1080，cuda10 Our detection is second level，also 1face . test1： === python test_kpoint.py [11:21:45] src/nnvm/legacy_json_util.cc:209: Loading symbol saved by previous version v1.3.0. Attempting to upgrade... [11:21:45] src/nnvm/legacy_json_util.cc:217: Symbol successfully upgraded!...
Thanks for your sharing. l have a problem when l want to deploy this to windows system, how to complie cython file in windows system? Can you help me, Thanks.
Thanks for your sharing I want to ask why you do "cv2.copyMakeBorder" before detect in test_kpoint.py? I have no idea. Could you teach me? Thanks.
hi thanks the great job. I have a problem when i wanna to inference the detector. follow is the error info: "`mxnet.base.MXNetError: [16:26:09] src/operator/nn/./cudnn/cudnn_convolution-inl.h:713: Check failed: e == CUDNN_STATUS_SUCCESS (2...
Dear author: I want to know what dataset and landmarks annotation form of you project, pleas tell me??
