AI Paper Atlas

Research map / 3D vision / Depth and stereo

ORCaS: Unsupervised Depth Completion via Occluded Region Completion as Supervision

Hyoungseob Park, Runjian Chen, Patrick Rim, Dong Lao, Alex Wong

ICLR 2026 · 3D vision › Depth and stereo · approach: Dense depth prediction and completion

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 ORCaS loss (feature supervision of occluded regions) on top of 2D-to-3D broadcast and 3D warping (ablation)ORCaS · VOID1500MAE: 40.52 -> 30.90Improved
add 3D warping without ORCaS loss (ablation)base model · VOID1500MAE: 35.31 -> 52.60Hurt
add 2D-to-3D broadcasting to base network (ablation)KBNet with transformer · VOID1500MAE: 35.31 -> 33.56Improved
add 3D broadcast to warping plus ORCaS loss (ablation)ORCaS · VOID1500MAE: 36.37 -> 30.90Improved
use feature supervision instead of depth supervision for occluded region (ablation)ORCaS · VOID1500MAE: 38.81 -> 30.90Improved
add feature supervision versus no supervision (ablation)ORCaS · VOID1500MAE: 33.56 -> 30.90Improved
increase depth planes from 2 to 8 (ablation)ORCaS · VOID1500MAE: 32.73 -> 30.90Improved
increase context pooling size from (2,2) to (4,4) (ablation)ORCaS · VOID1500MAE: 32.86 -> 30.90Improved
increase context pooling size from (4,4) to (8,8) (ablation)ORCaS · VOID1500MAE: 30.90 -> 31.36Hurt
mask moving objects during training (ablation)ORCaS · VOID1500MAE: 30.90 -> 30.84No clear effect
replace AugUndo with ORCaS for zero-shot transferAugUndo · ScanNet (zero-shot from VOID1500)MAE: 82.53 -> 68.86Improved
replace AugUndo with ORCaS under sparse inputAugUndo · VOID150MAE: 117.93 -> 81.89Improved

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