Research map / Vision-language and multimodal / Human and embodied modeling
What Do Deep Saliency Models Learn about Visual Attention?
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 |
|---|---|---|---|
| Use factorized features and probabilistic basis maps for saliency inference | DINet · OSIE | NSS: 2.88 -> 2.91 | Helped |
| Use factorized features and probabilistic basis maps for saliency inference | DINet · OSIE | CC: 0.63 -> 0.64 | No clear effect |
| Use factorized features and probabilistic basis maps for saliency inference | DINet · MIT | NSS: 2.54 -> 2.53 | Hurt |
| Use factorized features and probabilistic basis maps for saliency inference | DINet · MIT | CC: 2.70 -> 0.70 | No clear effect |
| Use factorized features and probabilistic basis maps for saliency inference | DINet · SALICON | NSS: 1.92 -> 1.89 | Hurt |
| Use factorized features and probabilistic basis maps for saliency inference | DINet · SALICON | CC: 0.87 -> 0.86 | Hurt |
| Fine-tune pretrained ImageNet features, especially during the first epoch | Saliency models · Saliency prediction | performance | Helped |
| Train on OSIE rather than SALICON to capture dataset-specific semantic weights | DINet · OSIE | social-cue semantic weight | Helped |
| Train on MIT rather than SALICON | DINet · MIT | vehicle semantic weight | Mixed |
| Use TranSalNet rather than the other tested architectures | Saliency models trained on SALICON · SALICON | clothing semantic weight | Helped |
| Train on attention data from people with autism rather than those without autism | DINet · Autism-group attention dataset | action-cue semantic weight | Hurt |
| Train on positive- or negative-emotion stimuli rather than neutral stimuli | DINet · EMoD | social- and action-cue semantic weight | Helped |
Closest papers by meaning
- Understanding and Visualizing Deep Visual Saliency Models (CVPR 2019 · similarity 0.93)
- A Deeper Look at Saliency: Feature Contrast, Semantics, and Beyond (CVPR 2016 · similarity 0.90)
- Learning From Unique Perspectives: User-Aware Saliency Modeling (CVPR 2023 · similarity 0.90)
- CaRDiff: Video Salient Object Ranking Chain of Thought Reasoning for Saliency Prediction with Diffusion (AAAI 2025 · similarity 0.88)
- Explainable Saliency: Articulating Reasoning with Contextual Prioritization (CVPR 2025 · similarity 0.87)
- Hallucination Begins Where Saliency Drops (ICLR 2026 · similarity 0.86)
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.