Pytorch_Retinaface
Pytorch_Retinaface copied to clipboard
biubug6
Retinaface get 80.99% in widerface hard val using mobilenet0.25.
请教一下 文章里面花了比较大的篇幅介绍了 dense 3D vertics 但是这个项目里面没有体现，可以说明一下为啥吗？（我猜是有没有这个loss对效果影响不大？） 多谢！
  
I'm movice, I test your trained model(mobilenet), it just get 73.82%, that means this model using orginal scale? if want to get the 80.99%, need to retrain this model again?...
Hello, can you give me the detection result using mobile0.25 on FDDB dataset (already convert to FDDB mode)?
The exported onnx model from this repo currently fails with TRT 7.x. The failure happens when the onnx model is parsed by TRT. The parsed model will be missing most...
Is the R50 result displayed on the page consistent with the training parameters of the mxnet version? I see that the training scale is 840 * 840 in R50‘s config.
Hi, Thank you very much for your materials. I have downloaded the weights from this link (https://drive.google.com/drive/folders/1oZRSG0ZegbVkVwUd8wUIQx8W7yfZ_ki1). But when I am running **python test_widerface.py --trained_model weight_file --network mobile0.25 or resnet50**...
As title, can I add negative images (without any annotations) to fine-tune the model?
