Research map / Detection and segmentation / CNN features
RGB-D Salient Object Detection with Cross-Modality Modulation and Selection
Chongyi Li, Runmin Cong, Yongri Piao, Qianqian Xu, Chen Change Loy
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 cross-modality feature modulation (cmFM) (ablation) | Ours (full RGB-D SOD network) · STEREO | F-measure: 0.8727 -> 0.9084 | Improved |
| add cross-modality feature modulation (cmFM) (ablation) | Ours (full RGB-D SOD network) · STEREO | MAE: 0.0722 -> 0.0422 | Improved |
| add cross-modality feature modulation (cmFM) (ablation) | Ours (full RGB-D SOD network) · DUT-Test | F-measure: 0.8968 -> 0.9328 | Improved |
| use cmFM instead of element-wise addition (cmFA) (ablation) | Ours (full RGB-D SOD network) · STEREO | F-measure: 0.9020 -> 0.9084 | Improved |
| use cmFM instead of concatenation (cmFC) (ablation) | Ours (full RGB-D SOD network) · STEREO | F-measure: 0.8995 -> 0.9084 | Improved |
| use cmFM instead of concatenation (cmFC) (ablation) | Ours (full RGB-D SOD network) · DUT-Test | MAE: 0.0617 -> 0.0366 | Improved |
| add adaptive feature selection (AFS) (ablation) | Ours (full RGB-D SOD network) · STEREO | F-measure: 0.8990 -> 0.9084 | Improved |
| add adaptive feature selection (AFS) (ablation) | Ours (full RGB-D SOD network) · DUT-Test | S-measure: 0.8666 -> 0.8853 | Improved |
| add fused spatial features (GFF) in AFS (ablation) | Ours (full RGB-D SOD network) · STEREO | MAE: 0.0690 -> 0.0422 | Improved |
| add channel attention-on-channel attention (CACA) (ablation) | Ours (full RGB-D SOD network) · STEREO | F-measure: 0.9017 -> 0.9084 | Improved |
| use AFS instead of conventional channel attention (CA) (ablation) | Ours (full RGB-D SOD network) · STEREO | S-measure: 0.8780 -> 0.8895 | Improved |
| add sg-PEA module (ablation) | Ours (full RGB-D SOD network) · STEREO | F-measure: 0.9057 -> 0.9084 | Improved |
Closest papers by meaning
- Deep RGB-D Saliency Detection With Depth-Sensitive Attention and Automatic Multi-Modal Fusion (CVPR 2021 · similarity 0.94)
- CNN-Based RGB-D Salient Object Detection: Learn, Select, and Fuse (IJCV 2021 · similarity 0.94)
- Specificity-Preserving RGB-D Saliency Detection (ICCV 2021 · similarity 0.93)
- Learning Selective Mutual Attention and Contrast for RGB-D Saliency Detection (TPAMI 2021 · similarity 0.93)
- Promoting Saliency From Depth: Deep Unsupervised RGB-D Saliency Detection (ICLR 2022 · similarity 0.93)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.90)
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.
