Research map / Detection and segmentation / CNN features
TENet: Triple Excitation Network for Video Salient Object Detection
Sucheng Ren, Chu Han, Xin Yang, Guoqiang Han, Shengfeng He
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 spatial excitation to temporal excitation (ablation) | TENet · DAVSOD | max F-beta: 0.591 -> 0.648 | Improved |
| add temporal excitation to spatial excitation (ablation) | TENet · DAVSOD | max F-beta: 0.615 -> 0.648 | Improved |
| add online excitation (1 iteration) | TENet · DAVSOD | max F-beta: 0.648 -> 0.659 | Improved |
| add online excitation (20 iterations) | TENet · DAVSOD | max F-beta: 0.648 -> 0.664 | Improved |
| use ground-truth saliency as excitation map (upper bound) (ablation) | TENet · DAVSOD | max F-beta: 0.648 -> 0.841 | Improved |
| add online excitation beyond 20 iterations (ablation) | TENet · DAVSOD | max F-beta | No clear effect |
| add GT-only curriculum learning (ablation) | Baseline · DAVSOD | max F-beta: 0.579 -> 0.584 | No clear effect |
| add GT-only curriculum learning (ablation) | Baseline · DAVSOD | convergence time: 80 hours -> 32 hours | Improved |
| add learned complementary excitation map (ablation) | Baseline · DAVSOD | max F-beta: 0.579 -> 0.641 | Improved |
| use semi-curriculum instead of learned excitation with pretraining (ablation) | Baseline + Learned Excitation · DAVSOD | convergence time: 46 (pre-training) + 38 hours -> 40 hours | Improved |
| add semi-curriculum learning (ablation) | Baseline · DAVSOD | max F-beta: 0.579 -> 0.648 | Improved |
| add excitation mechanism to training (ablation) | TENet · training | convergence time: 80 hours -> 40 hours | Improved |
Closest papers by meaning
- Motion Guided Attention for Video Salient Object Detection (ICCV 2019 · similarity 0.88)
- Pyramid Constrained Self-Attention Network for Fast Video Salient Object Detection (AAAI 2020 · similarity 0.87)
- Shifting More Attention to Video Salient Object Detection (CVPR 2019 · similarity 0.87)
- ViDSOD-100: A New Dataset and a Baseline Model for RGB-D Video Salient Object Detection (IJCV 2024 · similarity 0.86)
- Pixels, Regions, and Objects: Multiple Enhancement for Salient Object Detection (CVPR 2023 · similarity 0.86)
- What Is and What Is Not a Salient Object? Learning Salient Object Detector by Ensembling Linear Exemplar Regressors (CVPR 2017 · 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.
