-
### Description
As mentioned on Retinaface repo, Tinaface is light weight model which has better accuracy on wider {hard} dataset. and also it is efficient compared to retinaface. so could you add th…
-
Hello, I was wondering if there's a pretrained model for TinaFace.
I just want to see the quick output of the detected images, but it seems like there's no pretrained model.
Thank you!!
-
When I run the model on colab:
> !python tools/infer.py '/content/vedadet/configs/infer/tinaface/tinaface_r50_fpn_gn_dcn.py' '/content/vedadet/images/1.jpg'
I get the error:
> Traceback (mos…
-
Hi, First of all i wanna tank you for your great work.
i followed your instruction on Inference. first, i installed pytorch and cuda and then git clone the repo and then set root dir. after that i …
-
Hi,
Thank you for your great work!
I am reading your TinaFace paper and see that the outputs of TinaFace are in shape W x H x 4A. So I was wondering how do you convert those outputs into bounding b…
-
I am trying to reproduce the widerface results furnished here: https://github.com/Media-Smart/vedadet/tree/main/configs/trainval/tinaface.
I prepared the data as suggested in the `Data Preparation…
-
Hi, thanks for your share the interesting repo.
have you used the mobilev3 as backbone for tinaface?
whether the training strategy should be changed?
-
跑tinaface的评测，只有voc的ap，没有easy，medium，hard的结果：
+-------+-------+----------+--------+-------+
| class | gts | dets | recall | ap |
+-------+-------+----------+--------+-------+
| face | 319…
-
from eval_folder import Face_detection
#default size is 640 ----> No Problem
det = Face_detection(det_shape=640)
but...
det = Face_detection(det_shape=320)
is error code
[The det_shape size…
-
Environment info：
sys.platform: linux
Python: 3.7.13 (default, Oct 18 2022, 18:57:03) [GCC 11.2.0]
CUDA available: True
GPU 0,1: NVIDIA A100 80GB PCIe
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compi…