opensphere
opensphere copied to clipboard
ydwen
A hyperspherical face recognition library based on PyTorch
Hi, Thank you for sharing the code for SphereFace in Pytorch! Very interesting work! I have been using your code for SphereFace loss and was wondering how do you arrive...
Do you have a plan to release your trained SphereFace2 and SphereFace-R models for testing and reproducing paper numbers?
The **lw** in **SphereFace2** didn't appear in the paper, what does it meaning?
Hi, thank you for sharing your work! I'm wondering how to preprocess the original dataset to create the validation dataset? Is it a workflow about how to clean the dataset,...
Hello! What's the reason to skip gradient computation during normalization of classifier weights? ``` with torch.no_grad(): self.w.data = F.normalize(self.w.data, dim=0) ``` You've implemented all sphere losses in this way, so...
i want use sphereface2 in my face recognition project, could you please provide face recognition api or a tutorial of how to apply sphereface2 in project? thank you.
Hey @ydwen, Thanks for such an amazing work of SimPLE: Pairwise Similarity Learning and sharing the code for further research. It will be great if you help me understand below...
Hi guys, Kindly note that I have a single GPU and the below questions are asked based on the same i.e. my training is with a single GPU and not...
你好，我下载了reademe 中提供的预训练模型(sphereface2_vggface2_sfnet20、SphereFacePlus_ms1m_sfnet64bn)，放到代码中执行main_inference.py ，其中parse_args的设置如下 ```Python def parse_args(): parser = argparse.ArgumentParser( description='A PyTorch project for face recognition.') parser.add_argument('--image_dir', default='./face_alignment/test_images/', help='path to the image directory') parser.add_argument('--proj_dir', default='/workspace/SphereFacePlus_ms1m_sfnet64bn/20220429_154144', help='the directory to load model') args...
Hi Author, I am new to computer vision and particularly to face detection and recognition. I wonder if there is any example of implementation in using yolo to detect faces...
