peteryuX/retinaface-tf2
RetinaFace (Single-stage Dense Face Localisation in the Wild, 2019) implemented (ResNet50, MobileNetV2 trained on single GPU) in Tensorflow 2.0+. This is an unofficial implementation. With Colab.
PythonMIT
Issues
- 1
- 0
can't find the pretrained models
#54 opened by emnab - 2
- 5
how convert to .tflite?
#15 opened by xieshenru - 7
Error Test.py on one image
#25 opened by EsrtG - 0
- 5
ValueError: Received incompatible tensor with shape (1, 1, 32, 64) when attempting to restore variable with shape (1, 1, 192, 64) and name model/layer_with_weights-1/output1/conv/kernel/.ATTRIBUTES/VARIABLE_VALUE.
#44 opened by HamDan1999 - 0
- 1
- 1
Error Testing on WIDER FACE Validation Set
#41 opened by galuhputraa - 0
Access of widerface
#46 opened by Jmh0527 - 1
- 1
SSH context module
#39 opened by ilyanelken - 3
Error while running on tf2.4
#26 opened by harshithdwivedi - 0
frozen graph
#38 opened by fisakhan - 0
I'm wondering how to fine-tune the model, such as freezing the model, which levels can have a good training effect. Is there any research?
#36 opened by goldwater668 - 0
PiecewiseConstantWarmUpDecay incorrect implementation causes skipping to the almost minimal LR
#35 opened by or-toledano - 1
More than 5 landmarks
#28 opened by leopqrz - 1
test performance on big face
#32 opened by chenxinhua - 0
train size
#33 opened by alicera - 2
Very slow inference
#31 opened by fmmohammadi - 9
- 2
Inference with TF1.15
#17 opened by ilyanelken - 0
- 0
keras .h5 model
#29 opened by fisakhan - 0
inference time
#27 opened by AlexArtemis - 1
the question about the inference time
#21 opened by nick-zoo - 1
URL fetch failed
#8 opened by SURABHI-GUPTA - 9
tf.checkpoint how to convert to pb correct
#4 opened by EASY-EAI - 8
Can you add inference script?
#12 opened by abhiksark - 0
classification loss (crossentropy)
#23 opened by onlyNata - 1
CPU RAM consumption until depletion during training if shuffle is set to True
#18 opened by wingedrasengan927 - 0
- 0
Completely wrong prediction matrix!
#19 opened by zhjikoshlizhzc - 1
- 0
GPU utilisation is 0 while running test.py
#13 opened by bhatiyaarp - 0
support batching on test
#14 opened by ThiagoMateo - 2
about multi gpus
#10 opened by lovejing0306 - 1
Typo in the README
#9 opened by RaySuhyunLee - 2
hard negative mining
#6 opened by lovejing0306 - 2
- 1
Runtime performance
#2 opened - 1
Can this tf2 model be converted into pb model like in tensorflow 1.x and then be executed in tensorflow 1.x ?
#3 opened by hangdonglin - 4
Pre model download
#1 opened by tjulitianyi1997