AI Paper Atlas

Research map / Detection and segmentation / CNN features

Depth Coefficients for Depth Completion

Saif Imran, Yunfei Long, Xiaoming Liu, Daniel Morris

CVPR 2019 · Detection and segmentation › CNN features · approach: Domain-specific segmentation and detection · 114 citations (OpenAlex)

Check 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 Depth Coefficients with cross-entropy instead of sparse depth with MSE (ablation)Depth completion network · NYU2MAE: 6.63 -> 3.82Improved
Use Depth Coefficients input instead of sparse-depth input with MSE loss (ablation)Depth completion network · NYU2MAE: 6.63 -> 6.10Improved
Use cross-entropy instead of MSE with sparse-depth input (ablation)Depth completion network · NYU2MAE: 6.63 -> 9.53Hurt
Use 3-coefficient reconstruction instead of all coefficients (ablation)Depth completion network · KITTI, 16R LidarRMSE: 142.3 -> 160.6Hurt
Use 3-coefficient reconstruction instead of all coefficients (ablation)Depth completion network · KITTI, 16R LidartRMSE: 50.5 -> 47.2Improved
Use 3-coefficient reconstruction instead of all coefficients (ablation)Depth completion network · KITTI, 16R LidartMAE: 36.1 -> 33.4Improved
Use 3-coefficient reconstruction instead of all coefficients (ablation)Depth completion network · NYU2RMSE: 0.118 -> 0.131Hurt
Use 3-coefficient reconstruction instead of all coefficients (ablation)Depth completion network · NYU2tMAE: 0.042 -> 0.040Improved
Increase DC channels from 10 to 40 (ablation)Depth completion network · KITTI validation, 16R LidarMAE: 50.2 -> 37.8Improved
Increase DC channels from 40 to 80 (ablation)Depth completion network · KITTI validation, 16R LidarMAE: 37.8 -> 38.4Hurt
Upsample sparse 16-row Lidar with DC-3coeff instead of MaCar detection with Frustum PointNet · KITTI 3D bounding boxes, Easyaverage precision: 36.7 -> 64.9Improved
Upsample sparse 16-row Lidar with Ma instead of using raw sparse dataCar detection with Frustum PointNet · KITTI 3D bounding boxes, Easyaverage precision: 54.4 -> 36.7Hurt

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