insightface
insightface copied to clipboard
deepinsight
State-of-the-art 2D and 3D Face Analysis Project
当网络中使用到torch.split或torch.chunk的时候，会造成测试服务器CPU出现内存泄露并死机。 在本地服务器上测试发现当cudnn==8.0.5会造成这个问题，升级cudnn的版本可以解决这个问题。 麻烦升级 MFR测试服务器的cudnn版本来解决这个问题，多谢！
I get error of "Illegal Instruction (Core Dumped)" on trying to run the train.py in insightface/recognition/arcface_mxnet
Hello, I'm facing the following issue while trying to Fine-tune the Softmax model with Triplet loss: Command: CUDA_VISIBLE_DEVICES='0,1,2,3' python -u train.py --network m1 --loss triplet --lr 0.005 --pretrained ./models/m1-softmax-emore,1 Error:...
Why do we explicitly need to code the backward part in forward_backward in partial_fc.py for arcface_torch? can't PyTorch auto-differentiation take care of this? I would really appreciate help on this....
Not able to download the above mentioned model packs
hi, how to modify the tools/scrfd.py to enable support batch-size input data? and i have converted the scrfd.onnx with dynamic batch-size, but I can not get the correct post-process results...
how to apply adam and adamw optimizer to the partialfc training?
I did some modifications on the SCRFD model and its performance failed my expectation. I expanded 5 keypoints output to 68 and added a separate conv-bn module just like for...
1. why the classification loss have two value, i think one value is enough, such as 0 means no face, 1 means having face. 2. why there is a crop...
Metadata
Owner
Metadata
State-of-the-art 2D and 3D Face Analysis Project