Research map / Detection and segmentation / CNN features
Saliency Pattern Detection by Ranking Structured Trees
Lei Zhu, Haibin Ling, Jin Wu, Huiping Deng, Jin Liu
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add K-NN enhanced saliency propagation (ablation) | SSP prediction · DUT-OMRON | PR curve AUC: 0.7421 -> 0.7745 | Improved |
| add NN similarity to the adjacent-graph (beta 0 to 0.8) (ablation) | Saliency propagation graph · DUT-OMRON | PR curve AUC: 0.7584 -> 0.7745 | Improved |
| use NN similarity only versus adjacent similarity only (ablation) | Saliency propagation graph · DUT-OMRON | PR curve AUC: 0.7584 -> 0.7725 | Improved |
| add adjacent similarity to NN-only graph (beta 1 to 0.8) (ablation) | Saliency propagation graph · DUT-OMRON | PR curve AUC: 0.7725 -> 0.7745 | Improved |
| raise beta from 0.2 to 0.8 (ablation) | Saliency propagation graph · DUT-OMRON | PR curve AUC: 0.7662 -> 0.7745 | Improved |
| raise beta from 0.6 to 0.8 (ablation) | Saliency propagation graph · DUT-OMRON | PR curve AUC: 0.7733 -> 0.7745 | No clear effect |
| proposed method versus best deep baseline on PASCAL-S | RST · PASCAL-S | PR curve AUC: 0.8571 versus 0.8317 | Hurt |
| proposed method versus best deep baseline on SOD | RST · SOD | PR curve AUC: 0.8565 versus 0.8344 | Hurt |
| proposed method versus best deep baseline on ECSSD | RST · ECSSD | PR curve AUC: 0.9388 -> 0.9503 | Improved |
| proposed method versus best deep baseline on TCD | RST · TCD | PR curve AUC: 0.9247 -> 0.9409 | Improved |
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.
- RFCN — R-FCN: Object Detection via Region-based Fully Convolutional Networks (NeurIPS 2016) · better on 1 of 2 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
- Instance-Level Relative Saliency Ranking with Graph Reasoning (TPAMI 2021 · similarity 0.90)
- Salient Object Detection: A Discriminative Regional Feature Integration Approach (IJCV 2016 · similarity 0.90)
- Saliency Hierarchy Modeling via Generative Kernels for Salient Object Detection (ECCV 2022 · similarity 0.90)
- Structure-Consistent Weakly Supervised Salient Object Detection with Local Saliency Coherence (AAAI 2021 · similarity 0.89)
- Weakly-Supervised Salient Object Detection Using Point Supervision (AAAI 2022 · similarity 0.89)
- Ranking Saliency (TPAMI 2016 · similarity 0.89)
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.
