Depth Prediction Without the Sensors: Leveraging Structure for Unsupervised Learning from Monocular Videos

15 Nov 2018  ·  Vincent Casser, Soeren Pirk, Reza Mahjourian, Anelia Angelova ·

Models and examples built with TensorFlow

PDF Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Unsupervised Monocular Depth Estimation Cityscapes Struct2Depth RMSE 7.280 # 10
RMSE log 0.205 # 9
Square relative error (SqRel) 1.737 # 10
Absolute relative error (AbsRel) 0.145 # 10
Test frames 1 # 1
Monocular Depth Estimation KITTI Eigen split struct2depth absolute relative error 0.135 # 59

Methods


No methods listed for this paper. Add relevant methods here