-
Graph is on PASCAL
retinaface: resize = 1
retinaface-2: resize = 1.5 (or resize close to …
-
Thanks for your good work.
When I testing the Tf1 version on ubuntu 18.04 OS then pepa-face engine application some time detecting nonface things as a face like below image
, GPU computing, and the obj files.
1. I …
-
I want to use faceboxes to fine tune my dataset, does this project has pretrained model to finetune?
-
您好，这边有两个疑问:
一是face_detection的infer只用det0 = detect_face(image, shrink)，fps=5左右，是否可以提高
二是face_detection的infer目前只支持单张图像的infer，是否可以设置batchsize=10这样
-
I want to evaluate the model using CPU according to instructions in the README file. To do that, I first need to compile NMS module. But make.sh yields the following error:
```
$ ./make.sh
Traceb…
-
Whenever I try to import a model, the return is true but the graph has no nodes
I'm not sure if I'm doing something wrong. I'm able to run the model using python.
Tensorflow.Net 0.11.3 was a dire…
-
It seems that in FDDB evaluation, the input image must be upsampled by 3x in order to achieve the performance mentioned in the paper. This adds up the computational cost. Without upsampling, the perfo…
-
https://github.com/zisianw/FaceBoxes.PyTorch/blob/55b9d62cc6ec558b43e177eea1681a72ae43d05b/models/faceboxes.py#L144
I think this line should view with batch size the same as the training line.
-
I followed the instructions in the README and trained faceboxes against WIDER-FACES and after 300 epochs (14 hrs, Titan V) I end up with 98.55% vs. the pretrained model with 98.47%.
For Pascal, I g…