AI Paper Atlas

Research map / 3D vision / Depth and stereo

Distortion and Uncertainty Aware Loss for Panoramic Depth Completion

Zhiqiang Yan, Xiang Li, Kun Wang, Shuo Chen, Jun Li, Jian Yang

ICML 2023 · 3D vision › Depth and stereo · approach: Depth from light field and defocus

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
train with LDU loss instead of raw loss (ablation)M3PT · Matterport3DRMSE (mm): 138.9 -> 114.9Improved
train with LDU loss instead of raw loss (ablation)M3PT · Matterport3DMRE: 0.0164 -> 0.0162No clear effect
train with LDU loss instead of raw loss (ablation)GuideNet · Matterport3DRMSE (mm): 192.9 -> 164.8Improved
train with LDU loss instead of raw loss (ablation)HoHo-R · Matterport3DRMSE (mm): 199.2 -> 168.8Improved
train with LDU loss instead of raw loss (ablation)HoHo-R · 3D60RMSE (mm): 196.9 -> 168.3Improved
train with LDU loss instead of raw loss (ablation)GuideNet · 3D60RMSE (mm): 239.3 -> 192.3Improved
add distortion-aware loss LDA only (ablation)M3PT · Matterport3D and 3D60RMSE: 5.9mm and 8.4mm reductionImproved
add consistent uncertainty estimation LCUE on top of LUA (ablation)M3PT · Matterport3D and 3D60RMSE: 15.7mm and 19.4mm reductionImproved
combine all three losses (LDU) (ablation)M3PT · Matterport3D and 3D60RMSE: 23.9mm and 24.1mm reductionImproved
use uncertainty maps from convolution plus horizontal flip instead of convolution only (ablation)M3PT LCUE · Matterport3DRMSE (mm): 129.6 -> 123.2Improved
use vertical flip pair HF,VF instead of C,HF (ablation)M3PT LCUE · Matterport3DRMSE (mm): 123.2 -> 124.7Hurt
increase number of uncertainty maps from 2 to 4 (ablation)M3PT LCUE · Matterport3DRMSE (mm): 123.2 -> 123.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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