ScaleDepth: Decomposing Metric Depth Estimation into Scale Prediction and Relative Depth Estimation

Benchmark Model Rank Results
monocular-depth-estimation-on-ddadScaleDepth-NK#2RMSE: 6.097absolute relative error: 0.121Delta < 1.25: 0.871
monocular-depth-estimation-on-ibims-1ScaleDepth-NK#3RMSE: 0.59δ1.25: 0.778absolute relative error: 0.164
monocular-depth-estimation-on-kitti-eigenScaleDepth-K#14absolute relative error: 0.048RMSE: 1.987Sq Rel: 0.136
monocular-depth-estimation-on-nyu-depth-v2ScaleDepth-N#20absolute relative error: 0.074RMSE: 0.267log 10: 0.032
monocular-depth-estimation-on-sun-rgbdScaleDepth-NK#1Delta < 1.25: 0.866RMSE: 0.359absolute relative error: 0.129