Research map / Label-efficient and robust learning / Recognition and long-tail
End-To-End Saliency Mapping via Probability Distribution Prediction
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 |
|---|---|---|---|
| train the saliency predictor with Bhattacharyya distance loss instead of Euclidean loss (ablation) | PDP · SALICON validation | CC: 0.667 -> 0.740 | Helped |
| train the saliency predictor with Bhattacharyya distance loss instead of Euclidean loss (ablation) | PDP · SALICON validation | AUC-Judd: 0.865 -> 0.880 | Helped |
| train the saliency predictor with Bhattacharyya distance loss instead of Euclidean loss (ablation) | PDP · SALICON validation | sAUC: 0.761 -> 0.783 | Helped |
| train the saliency predictor with Bhattacharyya distance loss instead of Euclidean loss (ablation) | PDP · SALICON validation | NSS: 2.108 -> 2.419 | Helped |
| use a probability-distance loss rather than standard regression losses (ablation) | PDP · SALICON validation | CC: 0.667 -> 0.740 | Helped |
| use Bhattacharyya distance loss rather than KL-divergence loss (ablation) | PDP · SALICON validation | CC: 0.724 -> 0.740 | Helped |
| fine-tune the SALICON-trained predictor on VOCA-2012 (ablation) | PDP · VOCA-2012 | AUC: 0.845 -> 0.875 | Helped |
| fine-tune the SALICON-trained predictor on VOCA-2012 (ablation) | PDP · VOCA-2012 | KL: 7.92 -> 8.23 | Hurt |
| use PDP rather than UPC | saliency prediction · SALICON challenge | CC: 0.596 -> 0.765 | Helped |
| use PDP rather than DeepFix | saliency prediction · MIT-300 | CC: 0.78 -> 0.70 | Hurt |
| use PDP rather than LCQS-baseline | saliency prediction · OSIE | sAUC: 0.765 -> 0.797 | Helped |
Closest papers by meaning
- Predicting When Saliency Maps Are Accurate and Eye Fixations Consistent (CVPR 2016 · similarity 0.89)
- Sanity Checks for Saliency Maps (NeurIPS 2018 · similarity 0.88)
- Understanding and Visualizing Deep Visual Saliency Models (CVPR 2019 · similarity 0.88)
- CAMERAS: Enhanced Resolution and Sanity Preserving Class Activation Mapping for Image Saliency (CVPR 2021 · similarity 0.88)
- TempSAL - Uncovering Temporal Information for Deep Saliency Prediction (CVPR 2023 · similarity 0.87)
- What Do Deep Saliency Models Learn about Visual Attention? (NeurIPS 2023 · 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.