Research map / Detection and segmentation / CNN features
Recurrently Aggregating Deep Features for Salient Object Detection
Read the paperCheck 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 |
|---|---|---|---|
| recurrently aggregate MLIF with per-layer features (vs per-layer features only) (ablation) | RADF · DUT-OMRON | F-measure: 0.763 -> 0.789 | Helped |
| recurrent aggregation vs MLIF-only prediction (ablation) | RADF · DUT-OMRON | F-measure: 0.772 -> 0.789 | Helped |
| increase recurrent steps from 1 to 2 (ablation) | RADF · five saliency datasets | F-measure: 0.787 -> 0.789 (DUT-OMRON) | Helped |
| use separate instead of shared weights across recurrent steps (ablation) | RADF2 · five saliency datasets | F-measure: 0.793 -> 0.789 (DUT-OMRON); 0.911 -> 0.914 (HKU-IS) | Mixed |
| replace VGG backbone with DenseNet-161 (ablation) | RADF · ECSSD | F-measure: 0.924 -> 0.886 | Hurt |
| RADF vs best prior (DSS) | salient object detection · DUT-OMRON | F-measure: 0.771 -> 0.789 | Helped |
| RADF vs best prior (DSS) | salient object detection · SOD | F-measure: 0.842 -> 0.835 | Hurt |
Closest papers by meaning
- Progressive Attention Guided Recurrent Network for Salient Object Detection (CVPR 2018 · similarity 0.93)
- Deep Embedding Features for Salient Object Detection (AAAI 2019 · similarity 0.93)
- Salient Object Detection With Pyramid Attention and Salient Edges (CVPR 2019 · similarity 0.92)
- Non-Local Deep Features for Salient Object Detection (CVPR 2017 · similarity 0.91)
- A Mutual Learning Method for Salient Object Detection With Intertwined Multi-Supervision (CVPR 2019 · similarity 0.91)
- Cascaded Partial Decoder for Fast and Accurate Salient Object Detection (CVPR 2019 · similarity 0.91)
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.