Research map / Detection and segmentation / CNN features
Motion Guided Attention for Video Salient Object Detection
Haofeng Li, Guanqi Chen, Guanbin Li, Yizhou Yu
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 motion branch with MGA at both encoder and decoder (ablation) | appearance branch · FBMS | maxF: 0.867 -> 0.910 | Improved |
| add motion branch with MGA at both encoder and decoder (ablation) | appearance branch · DAVIS | maxF: 0.865 -> 0.902 | Improved |
| add MGA at encoder to dual branch with MGA-D (ablation) | dual branch+MGA-D · FBMS | maxF: 0.891 -> 0.910 | Improved |
| add MGA at decoder to dual branch with MGA-E (ablation) | dual branch+MGA-E · FBMS | maxF: 0.893 -> 0.910 | Improved |
| replace Add fusion with MGA modules (ablation) | Add · DAVIS | maxF: 0.864 -> 0.902 | Improved |
| replace Concat fusion with MGA modules (ablation) | Concat · FBMS | maxF: 0.822 -> 0.910 | Improved |
| replace Mul fusion with MGA modules (ablation) | Mul · FBMS | maxF: 0.810 -> 0.910 | Improved |
| use MGA-tm instead of Add at encoder (ablation) | E-Add · DAVIS | maxF: 0.876 -> 0.893 | Improved |
| use MGA-t instead of Add at encoder (ablation) | E-Add · FBMS | S-m: 0.890 -> 0.906 | Improved |
| use MGA-m instead of Add at decoder (ablation) | D-Add · FBMS | maxF: 0.902 -> 0.910 | Improved |
| add channel-wise attention (MGA-tmc vs MGA-tm) at encoder (ablation) | E-MGA-tm · DAVIS | maxF: 0.893 -> 0.902 | Improved |
| pretrain both branches on sub-tasks (T_ma vs T_a) (ablation) | T_a · DAVIS | maxF: 0.885 -> 0.902 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PiCANet — PiCANet: Learning Pixel-Wise Contextual Attention for Saliency Detection (CVPR 2018) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Flow Guided Recurrent Neural Encoder for Video Salient Object Detection (CVPR 2018 · similarity 0.91)
- Progressive Attention Guided Recurrent Network for Salient Object Detection (CVPR 2018 · similarity 0.90)
- Salient Object Detection With Pyramid Attention and Salient Edges (CVPR 2019 · similarity 0.89)
- Revisiting Salient Object Detection: Simultaneous Detection, Ranking, and Subitizing of Multiple Salient Objects (CVPR 2018 · similarity 0.89)
- A Mutual Learning Method for Salient Object Detection With Intertwined Multi-Supervision (CVPR 2019 · similarity 0.89)
- A Motion-aware Spatio-temporal Graph for Video Salient Object Ranking (NeurIPS 2024 · similarity 0.89)
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.
