FaceX-Zoo icon indicating copy to clipboard operation
FaceX-Zoo copied to clipboard

A PyTorch Toolbox for Face Recognition

Results 120 FaceX-Zoo issues
Sort by recently updated
recently updated
newest added

In class ImageDataset_SST, you set len by train_list. The issue there is in my dataset has only 8 classes. So to train an epoch so quickly. Should I return them...

感觉口罩人脸的识别模型对不戴口罩人脸的特征表达能力不高吧?作者有没有做过实验呢?

I am using the [repo](https://github.com/JDAI-CV/FaceX-Zoo/blob/main/addition_module/face_mask_adding/FMA-3D/README.md) to mask the MS1M-v2 dataset (from insightface) that already aligned and crop to 112x112. Do you have a code to generate mask face with the...

感谢你们的开源数据,这里我有一个问题,想请问一下,我想在RAF-DB数据集上做测试的话,我应该怎么重新训练呢?在介绍中只有AffectNet数据集的训练,所以我应该在哪些文件里做下修改呢?

Hi, There is no link to access `RepVGG` models on Google Drive. Would you please provide Google drive links for these models in [Experiments of SOTA backbones](https://github.com/JDAI-CV/FaceX-Zoo/tree/main/training_mode#31-experiments-of-sota-backbones) ?

Hi, In [`backbone/backbone_def.py#L93`](https://github.com/JDAI-CV/FaceX-Zoo/blob/7d17d2719ab4d473594be6f3fb497034d4e9dd56/backbone/backbone_def.py#L93), the function `AttentionNet()` is used which is not defined. Would you please clarify what this function is? Also which backbone did you mean by `backbone_type == 'AttentionNet_wj'`...

测试发现瓶颈在这个函数render_cy(),单张图片平均耗时要0.4s左右。。

Training Mode | LFW | CPLFW | CALFW | AgeDb | Models&Logs -- | -- | -- | -- | -- | -- Convention Training | 91.77 | 61.56 |...

I download some of your pretrained model that can perform face recognition, I found these model are initial with pretrain_model='mv_epoch_8.pt' before they are trained. So I want to train model...