Research map / Detection and segmentation / CNN features
An Iterative and Cooperative Top-Down and Bottom-Up Inference Network for Salient Object Detection
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 |
|---|---|---|---|
| Iteratively alternate top-down and bottom-up saliency inference (ablation) | Saliency inference network · DUTSTE | MAE: .062 -> .050 | Helped |
| Use recurrent rather than CNN inference layers (ablation) | VGGNet saliency inference network · SOD | MAE: .109 -> .099 | Helped |
| Use recurrent rather than CNN inference layers (ablation) | VGGNet saliency inference network · DUTSTE | MAE: .058 -> .050 | Helped |
| Use a ResNet backbone instead of VGGNet (ablation) | Proposed saliency inference network · SOD | F-score: .815 -> .820 | Helped |
| Increase RNN update steps from 1 to 2 (ablation) | VGGNet+convGRU · DUTSTE | F-score: .819 -> .830 | Helped |
| Increase RNN update steps from 2 to 3 (ablation) | VGGNet+convGRU · SOD | F-score: .815 -> .817 | Helped |
| Increase RNN update steps from 3 to 4 (ablation) | VGGNet+convGRU · SOD | F-score: .817 -> .812 | Hurt |
| Share weights across iterations for each layer (ablation) | VGGNet+convGRU · DUTSTE | MAE: .051 -> .050 | Helped |
| Share weights across layers and iterations within inference direction (ablation) | VGGNet+convGRU · DUTSTE | MAE: .050 -> .053 | Hurt |
| Share all top-down and bottom-up inference weights (ablation) | VGGNet+convGRU · DUTSTE | MAE: .050 -> .056 | Hurt |
| Add deeper supervision at inference iterations and RNN update steps (ablation) | VGGNet+convGRU · SOD | MAE: .106 -> .102 | Helped |
| Add deeper supervision at inference iterations and RNN update steps (ablation) | VGGNet+convGRU · DUTSTE | MAE: .054 -> .050 | Helped |
Closest papers by meaning
- Weakly Supervised Salient Object Detection Using Image Labels (AAAI 2018 · similarity 0.91)
- Salient Object Detection With Pyramid Attention and Salient Edges (CVPR 2019 · similarity 0.91)
- Revisiting Salient Object Detection: Simultaneous Detection, Ranking, and Subitizing of Multiple Salient Objects (CVPR 2018 · similarity 0.91)
- A Mutual Learning Method for Salient Object Detection With Intertwined Multi-Supervision (CVPR 2019 · similarity 0.91)
- Deep Embedding Features for Salient Object Detection (AAAI 2019 · similarity 0.91)
- Recurrently Aggregating Deep Features for Salient Object Detection (AAAI 2018 · 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.