AI Paper Atlas

Research map / Detection and segmentation / CNN features

Pattern-Affinitive Propagation Across Depth, Surface Normal and Semantic Segmentation

Zhenyu Zhang, Zhen Cui, Chunyan Xu, Yan Yan, Nicu Sebe, Jian Yang

CVPR 2019 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 339 citations (OpenAlex)

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
Jointly predict depth and normal with PAP rather than depth aloneNYUD-v2 depth estimation · NYUD-v2rmse: 0.570 -> 0.550Improved
Add cross-task propagation (ablation)PAP joint depth, segmentation, and normal model · NYUD-v2rmse: 0.574 -> 0.558Improved
Add the reconstruction network (ablation)PAP with cross-task propagation · NYUD-v2rmse: 0.558 -> 0.550Improved
Add pair-wise loss (ablation)PAP with cross-task propagation and reconstruction network · NYUD-v2rmse: 0.550 -> 0.543Improved
Learn the affinity matrix at a larger scale, from 1/16 to 1/8 input scale (ablation)PAP · NYUD-v2rmse: 0.543 -> 0.533Improved
Increase affinity-matrix scale from 1/8 to 1/4 input scale (ablation)PAP · NYUD-v2rmse: 0.533 -> 0.530Improved
Use L1 distance rather than inner product for similarity calculation (ablation)PAP · NYUD-v2rmse: 0.543 -> 0.540No clear effect
Jointly learn depth, segmentation, and normal rather than depth alone (ablation)PAP · NYUD-v2rmse: 0.570 -> 0.533Improved
Use PAP for depth estimationNYUD-v2 depth estimation · NYUD-v2rmse: 0.509 -> 0.497Improved
Use PAP for semantic segmentationNYUD-v2 semantic segmentation · NYUD-v2IoU: 50.2 -> 50.4Improved
Use PAP for semantic segmentationSUN-RGBD semantic segmentation · SUN-RGBDIoU: 50.3 -> 50.5Improved
Apply PAP rather than single-task predictionKITTI depth estimation · KITTI online benchmarkSILog: 14.58 -> 13.08Hurt

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