AI Paper Atlas

Research map / 3D vision / Depth and stereo

Non-Local Spatial Propagation Network for Depth Completion

Jinsun Park, Kyungdon Joo, Zhe Hu, Chi-Kuei Liu, In So Kweon

ECCV 2020 · 3D vision › Depth and stereo · approach: Dense depth prediction and completion · 396 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 non-local neighbors instead of fixed-local neighbors (ablation)NLSPN · KITTI DC validationRMSE (mm): 890.4 -> 884.1Improved
use non-local neighbors (abs-sum norm, with confidence) instead of fixed-local (ablation)NLSPN · KITTI DC validationRMSE (mm): 891.6 -> 889.5Improved
use non-local neighbors, depth variance of selected neighbors (ablation)NLSPN · KITTI DC validationdepth variance (mm): 22.7mm -> 11.6mmImproved
use learned affinity instead of color-based affinity (ablation)NLSPN · KITTI DC validationRMSE (mm): 930.3 -> 884.1Improved
add confidence to affinity normalization (ablation)NLSPN (gamma-learnable norm) · KITTI DC validationRMSE (mm): 891.3 -> 884.1Improved
add confidence to affinity normalization (ablation)NLSPN (fixed-local, abs-sum norm) · KITTI DC validationRMSE (mm): 908.4 -> 891.6Improved
add confidence to affinity normalization (ablation)NLSPN (non-local, abs-sum norm) · KITTI DC validationRMSE (mm): 903.1 -> 889.5Improved
use learnable gamma normalization instead of abs-sum norm (ablation)NLSPN · KITTI DC validationRMSE (mm): 889.5 -> 884.1Improved
use gamma-learnable norm instead of tanh-c (ablation)NLSPN · KITTI DC validationRMSE (mm): 886.4 -> 884.1Improved
use confidence-incorporated normalization instead of binary confidence mask (ablation)NLSPN · KITTI DC validationRMSE (mm): 892.9 -> 884.1Improved
use confidence-incorporated normalization instead of weighted summation (ablation)NLSPN · KITTI DC validationRMSE (mm): 884.8 -> 884.1Improved
relax normalization to abs-sum* instead of abs-sum (ablation)NLSPN · KITTI DC validationRMSE (mm): 889.5 -> 886.0Improved

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