AI Paper Atlas

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

Robust Depth Completion with Uncertainty-Driven Loss Functions

Yufan Zhu, Weisheng Dong, Leida Li, Jinjian Wu, Xin Li, Guangming Shi

AAAI 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Confidence calibration objectives · 40 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 uncertainty-driven loss with Jeffrey's prior and residual refinementDepth completion model · KITTI Depth Completion TestMAE: 199.59 -> 190.88Improved
Use uncertainty-driven loss with Jeffrey's prior and residual refinementDepth completion model · KITTI Depth Completion TestIMAE: 0.84 -> 0.83Improved
Use uncertainty-driven loss with Jeffrey's prior and residual refinementDepth completion model · KITTI Depth Completion TestIRMSE: 1.99 -> 1.98Improved
Use uncertainty-driven loss with Jeffrey's prior and residual refinementDepth completion model · KITTI Depth Completion TestRMSE: 741.68 -> 795.43Hurt
Add uncertainty-driven loss (ablation)Depth completion model · KITTI Depth Completion ValMAE: 211.01 -> 203.94Improved
Add uncertainty-driven loss (ablation)Depth completion model · KITTI Depth Completion ValRMSE: 792.34 -> 791.13No clear effect
Add Jeffrey's prior to uncertainty-driven loss (ablation)Depth completion model · KITTI Depth Completion ValRMSE: 791.13 -> 783.87Improved
Add residual learning (ablation)Uncertainty-driven depth completion model · KITTI Depth Completion ValMAE: 204.61 -> 186.90Improved
Add residual learning (ablation)Uncertainty-driven depth completion model · KITTI Depth Completion ValRMSE: 783.87 -> 833.67Hurt
Use 4 completion blocks rather than 1 (ablation)Multiscale Joint Prediction Model · KITTI Depth Completion ValMAE: 215.06 -> 204.61Improved
Use 4 completion blocks rather than 2 (ablation)Multiscale Joint Prediction Model · KITTI Depth Completion ValMAE: 205.95 -> 204.61Improved
Use 4 completion blocks rather than 3 (ablation)Multiscale Joint Prediction Model · KITTI Depth Completion ValMAE: 206.46 -> 204.61Improved

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