Research map / Detection and segmentation / CNN features
What Is and What Is Not a Salient Object? Learning Salient Object Detector by Ensembling Linear Exemplar Regressors
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use ELE+ instead of the strongest baseline | SOD detectors · SED1 | MAE: 0.134 -> 0.108 | Helped |
| Use ELE+ instead of the strongest baseline | SOD detectors · PASCAL-S | wFM: 0.527 -> 0.581 | Helped |
| Use ELE+ instead of the strongest baseline | SOD detectors · ECSSD | wFM: 0.578 -> 0.650 | Helped |
| Use ELE+ instead of the strongest baseline | SOD detectors · MSRA-B | MAE: 0.094 -> 0.071 | Helped |
| Use ELE+ instead of the strongest baseline | SOD detectors · THUR15K | wFM: 0.499 -> 0.567 | Helped |
| Use ELE+ instead of the strongest baseline | SOD detectors · XPIE | wFM: 0.560 -> 0.611 | Helped |
| Use the foregroundness map as input cue (ablation) | ELE on ECSSD · ECSSD | wFM: 0.602 vs. 0.578 (MST), 0.490 (RBD), 0.478 (MR) | Helped |
| Replace heuristic proposal features with 1024d GoogleNet features (ablation) | ELE on ECSSD · ECSSD | wFM: 0.649 -> 0.578 | Hurt |
| Replace enhancement-based fusion with maximum raw exemplar score (ablation) | ELE on ECSSD · ECSSD | wFM: 0.649 -> 0.540 | Hurt |
| Replace enhancement-based fusion with mean raw exemplar score (ablation) | ELE on ECSSD · ECSSD | wFM: 0.649 -> 0.588 | Hurt |
| Add negatives collected from all training images to each exemplar's negative bag (ablation) | ELE · SED1 | FM: 0.849 -> 0.854 | Helped |
| Add negatives collected from all training images to each exemplar's negative bag (ablation) | ELE · PASCAL-S | FM: 0.692 -> 0.710 | Helped |
Closest papers by meaning
- Revisiting Salient Object Detection: Simultaneous Detection, Ranking, and Subitizing of Multiple Salient Objects (CVPR 2018 · similarity 0.91)
- Deep Embedding Features for Salient Object Detection (AAAI 2019 · similarity 0.89)
- Non-Local Deep Features for Salient Object Detection (CVPR 2017 · similarity 0.89)
- Structure-Consistent Weakly Supervised Salient Object Detection with Local Saliency Coherence (AAAI 2021 · similarity 0.89)
- DVSOD: RGB-D Video Salient Object Detection (NeurIPS 2023 · similarity 0.89)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.87)
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.