Research map / Detection and segmentation / CNN features
Deep Contrast Learning 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 |
|---|---|---|---|
| Use the two-stream deep contrast model with CRF refinement instead of the strongest eligible baseline | DCL+ · MSRA-B test | maxF: 0.885 -> 0.916 | Helped |
| Use the two-stream deep contrast model with CRF refinement instead of FCN | DCL+ · HKU-IS | maxF: 0.867 -> 0.904 | Helped |
| Use the two-stream deep contrast model with CRF refinement instead of the strongest baseline | DCL+ · DUT-OMRON | maxF: 0.703 -> 0.757 | Helped |
| Use the two-stream deep contrast model with CRF refinement instead of MDF | DCL+ · PASCAL-S | maxF: 0.764 -> 0.822 | Helped |
| Use the two-stream deep contrast model with CRF refinement instead of FCN | DCL+ · SOD | maxF: 0.795 -> 0.832 | Helped |
| Add CRF refinement to DCL (ablation) | DCL · MSRA-B test | accuracy | Helped |
| Fuse the fully convolutional and segment-wise spatial pooling streams (ablation) | Deep contrast network · MSRA-B test | F-measure | Helped |
| Use multi-scale MS-FCN rather than only its last scale (ablation) | MS-FCN · MSRA-B test | F-measure | Helped |
| Use the fully convolutional stream rather than relying only on the segment-level stream (ablation) | Deep contrast network · MSRA-B test | F-measure | Helped |
| Set the number of superpixels between 200 and 300 rather than fewer than 200 or more than 300 (ablation) | Deep contrast network · Saliency datasets | performance | Helped |
Closest papers by meaning
- Salient Object Detection With Pyramid Attention and Salient Edges (CVPR 2019 · similarity 0.91)
- Non-Local Deep Features for Salient Object Detection (CVPR 2017 · similarity 0.90)
- Deep Level Sets for Salient Object Detection (CVPR 2017 · similarity 0.90)
- A Mutual Learning Method for Salient Object Detection With Intertwined Multi-Supervision (CVPR 2019 · similarity 0.90)
- Deep Embedding Features for Salient Object Detection (AAAI 2019 · similarity 0.90)
- Weakly Supervised Salient Object Detection Using Image Labels (AAAI 2018 · similarity 0.90)
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.