AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

IEBins: Iterative Elastic Bins for Monocular Depth Estimation

Shuwei Shao, Zhongcai Pei, Xingming Wu, Zhong Liu, Weihai Chen, Zhengguo Li

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 21 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use iterative elastic bins instead of the strongest listed baselineIEBins with Swin-Large · KITTI Eigen splitAbs Rel: 0.051 -> 0.050Improved
use IEBins instead of the strongest listed baselineIEBins · KITTI official testSILog: 10.14 -> 9.63Improved
use IEBins instead of the strongest listed baselineIEBins with Swin-Large · NYU-Depth-v2Abs Rel: 0.090 -> 0.087Improved
use IEBins instead of the strongest listed baselineIEBins with Swin-Large · SUN RGB-D zero-shotAbs Rel: 0.143 -> 0.135Improved
add elastic target bins to iterative bin refinement (ablation)Baseline + IBins · NYU-Depth-v2Abs Rel: 0.090 -> 0.087Improved
replace standard regression with iterative elastic bins (ablation)Baseline + Regression · NYU-Depth-v2Abs Rel: 0.095 -> 0.087Improved
use 4 bins instead of 16 (ablation)IEBins · NYU-Depth-v2Abs Rel: 0.087 -> 0.105Hurt
use 8 bins instead of 16 (ablation)IEBins · NYU-Depth-v2Abs Rel: 0.087 -> 0.089Hurt
use 32 bins instead of 16 (ablation)IEBins · NYU-Depth-v2Abs Rel: 0.087 -> 0.088No clear effect
use 4 bins instead of 16 (ablation)IEBins · KITTI Eigen splitAbs Rel: 0.050 -> 0.288Hurt
use 8 bins instead of 16 (ablation)IEBins · KITTI Eigen splitAbs Rel: 0.050 -> 0.054Hurt
use 32 bins instead of 16 (ablation)IEBins · KITTI Eigen splitAbs Rel: 0.050 -> 0.050No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy