RMSE: 0.523
Papers with Code · CC BY-SA 4.0
Depth Estimation
Depth Estimation is the task of measuring the distance of each pixel relative to the camera. Depth is extracted from either monocular (single) or stereo (multiple views of a scene) images. Traditional methods use multi-view geometry to find the relationship between the images. Newer methods can directly estimate depth by minimizing the regression loss, or by learning to generate a novel view from a sequence. The most popular benchmarks are KITTI and NYUv2. Models are typically evaluated according to a RMS metric.
<span class="description-source">Source: DIODE: A Dense Indoor and Outdoor DEpth Dataset </span>
논문
ORD: 0.3938D3R: 0.3222RMSE: 0.1598δ1.25: 0.6390
RMSE: 0.5468
RMSE: 0.635
RMSE: 0.496
RMSE: 0.509
RMSE: 0.536
RMSE: 0.465
RMSE: 0.565
RMSE: 0.5183
δ1.25: 0.969
RMSE: 0.538
absolute relative error: 0.093
RMSE: 0.586
absolute relative error: 0.0793absolute error: 0.197RMSE: 0.4279Delta < 1.25: 0.9376
RMSE: 0.497
RMSE: 0.641
RMSE: 0.565
RMSE: 0.530
RMSE: 0.59δ1.25: 0.778absolute relative error: 0.164
RMSE: 0.497
RMSE: 0.496
RMSE: 0.575
RMSE: 0.604
RMSE: 0.514
RMSE: 8.119