AI Paper Atlas

Research map / Detection and segmentation / CNN features

Interactive Segmentation As Gaussion Process Classification

Minghao Zhou, Hong Wang, Qian Zhao, Yuexiang Li, Yawen Huang, Deyu Meng, Yefeng Zheng

CVPR 2023 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation

Read the paper

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 GPCIS instead of RITMinteractive segmentation with ResNet50 · DAVISNoC@90: 6.05 -> 5.89Improved
Use GPCIS instead of RITMinteractive segmentation with ResNet50 · BerkeleyNoC@90: 2.95 -> 2.60Improved
Use GPCIS instead of FocusCutinteractive segmentation with ResNet50 · SBDNoC@85: 3.62 -> 3.80Hurt
Use GPCIS instead of FocusCutinteractive segmentation with ResNet50 · GrabCutNoC@90: 1.78 -> 1.82Hurt
Use GPCIS instead of RITMinteractive segmentation with ResNet50 · DAVISIoU&1: 72.89% -> 75.67%Improved
Use GPCIS instead of RITMinteractive segmentation with ResNet50 · DAVISSPC: 34.24 -> 38.82Hurt
Add LVI guidance (ablation)GPCIS · GrabCut, Berkeley, SBD, DAVIS (average)NoC@85: 2.98 -> 2.85Improved
Add LVI guidance (ablation)GPCIS · GrabCut, Berkeley, SBD, DAVIS (average)NoC@90: 4.07 -> 4.00Improved
Learn function-space kernel parameters with DKL-F (ablation)GPCIS · GrabCut, Berkeley, SBD, DAVIS (average)NoC@90: 4.16 -> 4.00Improved
Learn weight-space prior parameters with DKL-W (ablation)GPCIS · GrabCut, Berkeley, SBD, DAVIS (average)NoC@90: 4.34 -> 4.00Improved
Concatenate input RGB with deep features (ablation)GPCIS · GrabCut, Berkeley, SBD, DAVIS (average)NoC@90: 4.10 -> 4.00Improved
Reduce testing epsilon squared (ablation)GPCIS with ResNet50 · DAVISNoIC: 36 -> 30 -> 21 -> 15 ->15 -> 8 -> 2Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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