AI Paper Atlas

Research map / Point clouds / Shape and surface completion

On the Over-Smoothing Problem of CNN Based Disparity Estimation

Chuangrong Chen, Xiaozhi Chen, Hui Cheng

ICCV 2019 · Point clouds › Shape and surface completion · approach: Point cloud upsampling and normals · 62 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
replace full-band weighted average with single-modal weighted average at inference (ablation)PSMNet stack-hourglass (L_reg) · Sceneflow3px SEE5: 9.40 -> 4.17Improved
replace full-band with single-modal weighted average at inference (ablation)PSMNet basic (L_reg) · Sceneflow3px SEE5: 10.97 -> 5.36Improved
replace full-band with single-modal weighted average at inference (ablation)GCNet (L_reg) · Sceneflow3px SEE5: 9.81 -> 5.39Improved
replace full-band with single-modal weighted average at inference (ablation)PDSNet (L_reg) · Sceneflow3px SEE5: 12.11 -> 6.01Improved
replace full-band with single-modal weighted average at inference (ablation)PSMNet stack-hourglass (L_reg) · Middlebury3px SEE5: 23.97 -> 17.94Improved
train with Gaussian cross entropy instead of smooth L1 regression loss (single-modal inference) (ablation)PSMNet stack-hourglass · Sceneflow3px SEE5: 4.17 -> 2.53Improved
train with Gaussian cross entropy instead of smooth L1 regression loss (single-modal inference) (ablation)PSMNet stack-hourglass · Sceneflow3px EPE: 2.83 -> 2.21Improved
use Gaussian instead of Laplace ground-truth distribution in cross entropy (ablation)GCNet · Sceneflow3px SEE5: 6.91 -> 3.92Improved
use Gaussian instead of Laplace ground-truth distribution in cross entropy (ablation)PDSNet · Sceneflow3px SEE5: 10.04 -> 3.43Improved
use Laplace cross entropy instead of smooth L1 loss (ablation)GCNet (single-modal) · Sceneflow3px SEE5: 5.39 -> 6.91Hurt
use Laplace cross entropy instead of smooth L1 loss (ablation)PDSNet (single-modal) · Sceneflow3px SEE5: 6.01 -> 10.04Hurt
lower the variance of the Gaussian ground-truth distribution (ablation)PSMNet stack-hourglass · Sceneflow3px SEE5Improved

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