Research map / Detection and segmentation / CNN features
M4-SAM: Multi-Modal Mixture-of-Experts with Memory-Augmented SAM for RGB-D Video Salient Object Detection
Jiyuan Liu, Jia Lin, Xiaofei Zhou, Runmin Cong, Deyang Liu, Zhi Liu
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 actual depth instead of RGB copy as pseudo depth (ablation) | M4-SAM · RDVS | E-measure: 0.858 -> 0.927 | Improved |
| use actual depth instead of all-zero depth (ablation) | M4-SAM · RDVS | E-measure: 0.895 -> 0.927 | Improved |
| fine-tune encoder with Modality-Aware MoE-LoRA vs frozen (ablation) | M4-SAM · RDVS | E-measure: 0.839 -> 0.927 | Improved |
| Modality-Aware MoE-LoRA vs vanilla LoRA (ablation) | M4-SAM · RDVS | E-measure: 0.904 -> 0.927 | Improved |
| Modality-Aware MoE-LoRA vs LoRA training memory (ablation) | M4-SAM · RDVS | GPU memory (GB): 14.25 -> 11.62 | Improved |
| Conv-LoRA instead of vanilla LoRA (ablation) | SAM2 encoder · RDVS | E-measure: 0.904 -> 0.900 | No clear effect |
| MoE top-K 2 instead of 1 (ablation) | M4-SAM · RDVS | E-measure: 0.903 -> 0.927 | Improved |
| MoE top-K 2 instead of 3 (ablation) | M4-SAM · RDVS | E-measure: 0.914 -> 0.927 | Improved |
| add memory bank to baseline (ablation) | M4-SAM · RDVS | E-measure: 0.908 -> 0.910 | No clear effect |
| add Gated Multi-Level Feature Fusion to memory model (ablation) | M4-SAM · RDVS | E-measure: 0.910 -> 0.927 | Improved |
| use mid-level decoder feature X_D^2 instead of X_D^1 (ablation) | M4-SAM · RDVS | E-measure: 0.918 -> 0.927 | Improved |
| use both X_D^1 and X_D^2 (ablation) | M4-SAM · RDVS | E-measure: 0.927 -> 0.909 | Hurt |
Closest papers by meaning
- SAM-DAQ: Segment Anything Model with Depth-guided Adaptive Queries for RGB-D Video Salient Object Detection (AAAI 2026 · similarity 0.93)
- ViDSOD-100: A New Dataset and a Baseline Model for RGB-D Video Salient Object Detection (IJCV 2024 · similarity 0.91)
- RA-SOD: Reliability-Aware RGB-T Salient Object Detection under Modality Degradation (ECCV 2026 · similarity 0.90)
- Siamese Network for RGB-D Salient Object Detection and Beyond (TPAMI 2021 · similarity 0.90)
- SAM-I2V: Upgrading SAM to Support Promptable Video Segmentation with Less than 0.2% Training Cost (CVPR 2025 · similarity 0.88)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.88)
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.
