FeatherNets_Face-Anti-spoofing-Attack-Detection-Challenge-CVPR2019

FeatherNets_Face-Anti-spoofing-Attack-Detection-Challenge-CVPR2019 copied to clipboard

过拟合问题

3

你好，我按照原代码的设置，用FeatherNetB在CASIA_SURF的原始数据集的深度图上进行训练和验证的时候发现训练集的accuracy和loss曲线是正常的，验证集上的accuracy是先升后降，loss是先降后升的，但是best模型的ACER值又能达到论文上面的精度，请问这个问题该怎么处理？

LiangXiaoEr

CASIA-SURF Dataset能分享下吗？

12

Adhders

有關RGB-D-IR的問題

4

您好，在 RGB-D-NI數據集中，深度圖與近紅外圖的分辨率是一樣的，但是可見光的分辨率與近紅外和深度圖的分辨率都不一樣，這3種圖像該怎麼融合呢？您這邊是分別預測的嗎

zcl912

Demo script to classify one image

6

Hello, could you provide a simple script to classify a single image? Really confused about the 1024 vector feature map, and all other issues are in chinese, so I dont...

kadirbeytorun

如何预测一张图片?

1

你好，请问如何使用你们的预训练模型预测一张人脸图片是否为真人，有相关的介绍或者预测脚本吗?

zhouhao-learning

RuntimeError: Error(s) in loading state_dict for DataParallel: size mismatch for module.fish.fish.9.4.1.weight: copying a param with shape torch.Size([1056]) from checkpoint, the shape in current model is torch.Size([1000, 1056, 1, 1]). size...

Jugerr

``` train_dataset = CASIA( transforms.Compose([ transforms.RandomResizedCrop(img_size), transforms.RandomHorizontalFlip(), transforms.ToTensor(), ColorAugmentation(), normalize, ]),phase_train=True) val_dataset = CASIA( transforms.Compose([ transforms.Resize(int(256 * ratio)), transforms.CenterCrop(img_size), transforms.ToTensor(), normalize, ]),phase_train=False,phase_test=args.phase_test) ```

Melonia-H

nohup python main.py --config="cfgs/fishnet150-32.yaml" --b 32 --lr 0.01 --every-decay 30 --fl- gamma 2 >> fishnet150-train.log Traceback (most recent call last): File "main.py", line 406, in main() File "main.py", line 113,...

marcoippolito

Did you train with only RGB image?

1

Hi @SoftwareGift If you trained model with only RGB image, please provide me the results. Thanks for your contributions.

ngoanpv

如何对任意一张图片进行活体检测？人脸检测，拿出人脸直接输入？？？或者这个人脸需要怎么处理？？

4

jiagh2010