Research map / Detection and segmentation / Open-set and few-shot detection
Empowering DINO Representations for Underwater Instance Segmentation via Aligner and Prompter
Zhiyang Chen, Chen Zhang, Hao Fang, Runmin Cong
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 AquaStyle Aligner to frozen DINOv2 + Mask2Former (ablation) | DiveSeg · UIIS test | mask mAP: 30.9 -> 34.1 (Table 4); full model 34.1 -> 35.6 | Improved |
| add ObjectPrior Prompter (ablation) | DiveSeg · UIIS test | mask mAP: 34.8 -> 35.6 | Improved |
| add AquaStyle Aligner when Prompter is present (ablation) | DiveSeg · UIIS test | mask mAP: 34.8 -> 35.6 | Improved |
| full fine-tuning of DINOv2 instead of frozen (ablation) | DINOv2+Mask2Former · UIIS test | mask mAP: 30.9 -> 31.1 | No clear effect |
| add LoRA to frozen DINOv2 (ablation) | DINOv2+Mask2Former · UIIS test | mask mAP: 30.9 -> 31.8 | Improved |
| add standard Adapter to frozen DINOv2 (ablation) | DINOv2+Mask2Former · UIIS test | mask mAP: 30.9 -> 32.7 | Improved |
| AquaStyle Aligner instead of standard Adapter (ablation) | DINOv2+Mask2Former · UIIS test | mask mAP: 32.7 -> 34.1 | Improved |
| replace USIS-SAM with DiveSeg (DINOv2 ViT-L) | underwater instance segmentation · USIS10K (multi-class) | mask mAP: 43.1 -> 48.4 | Improved |
| replace USIS-SAM with DiveSeg (DINOv2 ViT-L) | underwater instance segmentation · UIIS test | mask mAP: 29.4 -> 35.6 | Improved |
Closest papers by meaning
- Diving into Underwater: Segment Anything Model Guided Underwater Salient Instance Segmentation and A Large-scale Dataset (ICML 2024 · similarity 0.89)
- NAUTILUS: A Large Multimodal Model for Underwater Scene Understanding (NeurIPS 2025 · similarity 0.85)
- Fantastic Animals and Where to Find Them: Segment Any Marine Animal with Dual SAM (CVPR 2024 · similarity 0.85)
- A Unified Image-Dense Annotation Generation Model for Underwater Scenes (CVPR 2025 · similarity 0.85)
- SeafloorAI: A Large-scale Vision-Language Dataset for Seafloor Geological Survey (NeurIPS 2024 · similarity 0.85)
- Atlantis: Enabling Underwater Depth Estimation with Stable Diffusion (CVPR 2024 · similarity 0.83)
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.
