% < 11.25: 69.7% < 22.5: 76.2% < 30: 78.8Mean: 19.6
Papers with Code · CC BY-SA 4.0
Surface Normals Estimation
Surface normal estimation deals with the task of predicting the surface orientation of the objects present inside a scene. Refer to Designing Deep Networks for Surface Normal Estimation (Wang et al.) to get a good overview of several design choices that led to the development of a CNN-based surface normal estimator.
논문
% < 11.25: 71.1% < 22.5: 85.4% < 30: 89.8Mean Angle Error: 11.8
% < 11.25: 69.9Mean: 15.8
% < 11.25: 50.9% < 22.5: 65.2% < 30: 70Mean Angle Error: 28
RMSE : 10.76
RMSE : 11.84
RMSE : 11.8
Cosine Distance: 0.06
RMSE : 10.26
RMSE : 10.11
% < 11.25: 63.8% < 22.5: 79.8% < 30: 85.6Mean Angle Error: 14.6
Mean Angle Error: 14.15
RMSE : 9.76
RMSE : 10.22
Cosine Distance: 0.06
Cosine Distance: 0.28
Cosine Distance: 0.29
RMSE : 12.41
Cosine Distance: 0.06
RMSE: 12.2
Cosine Distance: 0.17
% < 11.25: 65.66% < 22.5: 82.28% < 30: 87.83Mean Angle Error: 13.09
L1 error: 4.80
Cosine Distance: 0.04