Fa-Ting Hong

追踪框存在少许飘移

1

在做追踪的时候，bbox会在画面中存在前后帧飘移现象，但如果去掉跟踪，bbox是不会飘移的，说明检测器检测出来的bbox很好，应该是做跟踪的时候对bbox做了微调，请问是在哪做的？

Could you release the code that computes these metrics？

关于ocr的模型

4

 请问这个h5文件没有下载地址吗

Hi @yoyo-nb , Thanks for your excellent work. The [video-processing](https://github.com/AliaksandrSiarohin/video-preprocessing) provides a large test set for VoxCeleb. Do you use all samples in that test set? If not, could you...

Hi, When I install version 0.3.7, and run: `out = detector.detect_image(img_path) gt_aus = out.facepose().values` an error was called:  But this interface was available half a year ago.

Process freeze!

1

Hi, Thanks for your nice work. After I install StyleHEAT and run the command: `CUDA_VISIBLE_DEVICES=0 python inference.py --config configs/inference.yaml --video_source=./docs/demo/videos/RD_Radio34_003_512.mp4 -output_dir=./docs/demo/output --if_extract` The process freezes as shown in the following...

How to obtain the uNet_test.npy if I train on a new dataset.

2

Hi, I find that the ''./data/uNet_test.npy'' is not produced by your code, and how can i obtain it if I apply this repo on a new dataset?

Thanks for your nice work. Could you please share the training log on the Voxceleb dataset. I trained my own model on voxceleb and would like to compare with your...

How to draw this picture using the face_result['lmks3d']?

关于数据集

3

请问你的PISC是怎么解压的？还有你的PIPA-relation的数据集有没有下载路径，那链接只有PIPA原始数据集的下载方式