protossw512/AdaptiveWingLoss
[ICCV 2019] Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression - Official Implementation
PythonApache-2.0
Issues
- 2
how visualize result?
#17 opened by yongyongdown - 3
- 4
cant find AdaptiveWingLoss implement
#10 opened by dtiny - 0
- 0
NME calculation for 68 points
#29 opened by hyperparameters - 4
- 0
- 0
Get 4.68 NME on WFLW dataset
#27 opened by Suke0 - 8
Training code implementation
#14 opened by HassanAbbas92 - 1
Please consider releasing the training code
#24 opened by noranart - 2
About the model's generalization ability
#20 opened by HongRealLee - 5
- 0
Run the code on custom data?
#23 opened by antithing - 0
why facial landmarks for covered parts?
#22 opened by sri9s - 0
recently can u realese the training code
#21 opened by jiangtaoo2333 - 1
68 Facial landmarks model
#16 opened by dguari1 - 2
what is output dimention?
#15 opened by aligoglos - 7
inference speed
#7 opened by dingweichao123 - 1
- 3
How to show boundary map
#11 opened by a462428 - 1
- 3
- 5
The error maybe be about the model
#8 opened by longlongvip - 1
Usage of coordconv?
#5 opened by PlumedSerpent - 1
- 3
Does the 'GTBbox' mean the bbox generated by MTCNN or the coordinates in annotation?
#3 opened by PlumedSerpent - 1
convert_WFLW.py script missing?
#2 opened by PlumedSerpent - 4