-
在test_model.py 中，我找到如下代码
```
landmark_L1 = graph.get_tensor_by_name('landmark_L1:0')
landmark_L2 = graph.get_tensor_by_name('landmark_L2:0')
landmark_L3 = graph.get_tensor_by_name('landmark_L3:0')…
-
**line 298:**
#14 * 14 * 128
conv5_1 = slim.convolution2d(conv4_1, 512, [1, 1], stride=2, activation_fn=None,scope='conv5_1/expand')
**line 362:**
#14 * 14 * 128
conv6_…
-
https://github.com/guoqiangqi/PFLD/blob/17363eae0885dcd6ff5ebe89bba2f8a991b72f34/model2.py#L28
-
误差非常大
-
@guoqiangqi OK, I see it. You has calculated it by adding the GT of attribute classes. Thanks!
_Originally posted by @briantse100 in https://github.com/guoqiangqi/PFLD/issues/1#issuecomment-4982097…
-
Hi，
I found this code is for training on 98-pts dataset. Do you have trained on 68-pts dataset 300-W and get the same accuracy as the PFLD paper said?
-
According to my understanding, it seems we can calculate the Euler Angle GT and w_n of all samples, and return them in dataset iterators. Right?
-
Hi， thanks for your open source, I tried to training model, but is not good to used, can we provide an available model in this project? Thanks @guoqiangqi
-
请问，《PFLD:A Practical Facial Landmark Detector》是你们的论文吗？
-