Research map / Detection and segmentation / CNN features
Siamese Network for RGB-D Salient Object Detection and Beyond
Keren Fu, Deng-Ping Fan, Ge-Peng Ji, Qijun Zhao, Jianbing Shen, Ce Zhu
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 |
|---|---|---|---|
| use CM cooperative fusion instead of channel concatenation (ablation) | JL-DCF · RGBD135 | S-measure: 0.916 -> 0.929 | Improved |
| add depth to RGB-only input (ablation) | JL-DCF (ResNet+RGB) · NJU2K | S-measure: 0.895 -> 0.903 | Improved |
| add depth to RGB-only input (ablation) | JL-DCF (ResNet+RGB) · STERE | S-measure: 0.909 -> 0.905 | Hurt |
| add RGB to depth-only input (ablation) | JL-DCF (ResNet+D) · STERE | S-measure: 0.744 -> 0.905 | Improved |
| use joint learning with a shared backbone instead of separate backbones (ablation) | SL-DCF (VGG+CM+RGB-D) · NJU2K | S-measure: 0.886 -> 0.897 | Improved |
| use ResNet-101 instead of VGG-16 backbone (ablation) | JL-DCF · NJU2K | S-measure: 0.897 -> 0.903 | Improved |
| add FA aggregation modules to decoder (ablation) | JL-DCF · NJU2K | S-measure: 0.893 -> 0.903 | Improved |
| add dense connections to decoder (ablation) | JL-DCF · LFSD | S-measure: 0.854 -> 0.862 | Improved |
| replace dense connections by a residual connection from FA5 to FA1 (ablation) | JL-DCF · NJU2K | S-measure: 0.902 -> 0.904 | No clear effect |
| replace ResNet-101 with DenseNet-161 backbone (ablation) | JL-DCF · NJU2K | S-measure: 0.903 -> 0.917 | Improved |
| co-train with extra RGB SOD data (JL-DCF*) | JL-DCF · SIP | S-measure: 0.879 -> 0.892 | Improved |
| co-train with extra RGB SOD data (JL-DCF*) | JL-DCF · VT821 (RGB-T SOD) | S-measure: 0.873 -> 0.885 | Improved |
Closest papers by meaning
- JL-DCF: Joint Learning and Densely-Cooperative Fusion Framework for RGB-D Salient Object Detection (CVPR 2020 · similarity 0.95)
- RGB-D Salient Object Detection with Cross-Modality Modulation and Selection (ECCV 2020 · similarity 0.92)
- Specificity-Preserving RGB-D Saliency Detection (ICCV 2021 · similarity 0.92)
- Promoting Saliency From Depth: Deep Unsupervised RGB-D Saliency Detection (ICLR 2022 · similarity 0.92)
- Accurate RGB-D Salient Object Detection via Collaborative Learning (ECCV 2020 · similarity 0.92)
- Progressively Complementarity-Aware Fusion Network for RGB-D Salient Object Detection (CVPR 2018 · similarity 0.92)
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.
