AI Paper Atlas

Research map / 3D vision / Depth and stereo

Towards Zero-Shot Scale-Aware Monocular Depth Estimation

Vitor Guizilini, Igor Vasiljevic, Dian Chen, Rareș Ambruș, Adrien Gaidon

ICCV 2023 · 3D vision › Depth and stereo · approach: Self-supervised monocular depth · 122 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
add input-level geometric embeddings (camera rays) to DPT (ablation)DPT · KITTI zero-shot, no median scalingAbsRel: 0.240 -> 0.144Improved
use 3D geometric embeddings instead of 2D positional embeddings (ablation)ZeroDepth · KITTI zero-shot, no median scalingAbsRel: 0.175 -> 0.100Improved
add intermediate variational latent representation on top of geometric-embedded DPT (ablation)ZeroDepth · KITTI zero-shot, no median scalingAbsRel: 0.144 -> 0.100Improved
use variational latent representation instead of deterministic Perceiver latent (ablation)ZeroDepth · KITTI zero-shot, no median scalingAbsRel: 0.122 -> 0.100Improved
add resolution jittering augmentation (ablation)ZeroDepth · KITTI zero-shot, no median scalingAbsRel: 0.142 -> 0.100Improved
add ray jittering augmentation (ablation)ZeroDepth · KITTI zero-shot, no median scalingAbsRel: 0.112 -> 0.100Improved
add embedding dropout (ablation)ZeroDepth · KITTI zero-shot, no median scalingAbsRel: 0.109 -> 0.100Improved
add surface normal regularization (ablation)ZeroDepth · KITTI zero-shot, no median scalingAbsRel: 0.109 -> 0.100Improved
train a single model on both indoor and outdoor data instead of separate models (ablation)ZeroDepth · KITTI outdoor vs NYUv2 indoorAbsRel: KITTI 0.100 -> 0.102; NYUv2 (metric) 0.104 -> 0.100Mixed
remove Waymo or LSD real-world training data (ablation)ZeroDepth · KITTI zero-shot, no median scalingAbsRel: 0.100 -> 0.133 (without LSD)Hurt
keep only the 50% lowest-uncertainty pixels using 50 samplesZeroDepth · KITTI zero-shotRMSE: 4.044 -> 1.678Improved
reduce latent dimension D_l to 32 (ablation)ZeroDepth · KITTIRMSE (metric): 4.904 (median-scaled), 6.421 (metric)Hurt

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