Research map / Video / Video generation
Unified Image and Video Saliency Modeling
Richard Droste, Jianbo Jiao, J. Alison Noble
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add domain-adaptive fusion (DA-Fusion) (ablation) | UNISAL baseline with smoothing · DHF1K val | CC: 0.416 -> 0.434 | Improved |
| add domain-adaptive fusion (DA-Fusion) (ablation) | UNISAL baseline with smoothing · SALICON val | CC: 0.799 -> 0.820 | Improved |
| add domain-adaptive batch normalization (DABN) (ablation) | UNISAL baseline with smoothing · DHF1K val | CC: 0.416 -> 0.396 | Hurt |
| add domain-adaptive batch normalization (DABN) (ablation) | UNISAL baseline with smoothing · SALICON val | CC: 0.799 -> 0.807 | Improved |
| add domain-adaptive Gaussians (ablation) | UNISAL baseline with smoothing · DHF1K val | NSS: 2.305 -> 2.278 | Hurt |
| add domain-adaptive Gaussians (ablation) | UNISAL baseline with smoothing · SALICON val | NSS: 1.654 -> 1.647 | No clear effect |
| add domain-adaptive smoothing (ablation) | UNISAL baseline with smoothing · DHF1K val | NSS: 2.305 -> 2.302 | No clear effect |
| add domain-adaptive smoothing (ablation) | UNISAL baseline with smoothing · SALICON val | CC: 0.799 -> 0.793 | Hurt |
| add Bypass RNN (ablation) | UNISAL baseline with smoothing · SALICON val | CC: 0.799 -> 0.819 | Improved |
| add Bypass RNN (ablation) | UNISAL baseline with smoothing · DHF1K val | CC: 0.416 -> 0.412 | Hurt |
| add image dataset SALICON to joint video training (setting v to vi) (ablation) | UNISAL · SALICON test | KLD: 0.48 -> 0.35 | Improved |
| train on video datasets only instead of SALICON (setting v vs iv) (ablation) | UNISAL · SALICON test | CC: 0.88 -> 0.74 | Hurt |
Closest papers by meaning
- Saliency Detection with Recurrent Fully Convolutional Networks (ECCV 2016 · similarity 0.87)
- Revisiting Video Saliency: A Large-Scale Benchmark and a New Model (CVPR 2018 · similarity 0.86)
- Saliency-Aware Video Object Segmentation (TPAMI 2017 · similarity 0.85)
- GraB: Visual Saliency via Novel Graph Model and Background Priors (CVPR 2016 · similarity 0.85)
- A Motion-aware Spatio-temporal Graph for Video Salient Object Ranking (NeurIPS 2024 · similarity 0.83)
- Unsupervised Salient Instance Detection (CVPR 2024 · similarity 0.82)
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.
