Research map / Detection and segmentation / 3D detection and driving
RGB-D Indiscernible Object Counting in Underwater Scenes
Guolei Sun, Xiaogang Cheng, Zhaochong An, Xiaokang Wang, Yun Liu, Deng-Ping Fan, Ming-Ming Cheng, Luc Van Gool
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 regression branch to density branch (ablation) | IOCFormer · IOCfish5K val (500) | MAE: 18.25 -> 16.94 | Improved |
| add density branch to regression branch (ablation) | IOCFormer · IOCfish5K val (500) | MAE: 17.47 -> 16.94 | Improved |
| add density-enhanced transformer encoder (DETE) (ablation) | IOCFormer · IOCfish5K val (500) | MAE: 16.94 -> 15.91 | Improved |
| add density-enhanced transformer encoder (DETE) (ablation) | IOCFormer · IOCfish5K val (500) | MSE: 35.92 -> 34.08 | Improved |
| increase DETE layers L from 2 to 8 (ablation) | IOCFormer · IOCfish5K val (500) | MAE: 16.75 -> 15.72 | Improved |
| increase DETE layers L from 6 to 8 (ablation) | IOCFormer · IOCfish5K val (500) | MAE: 15.91 -> 15.72 | No clear effect |
| add cross-agent attention interactions (L_m=2) instead of plain addition (ablation) | IOCFormer-D · IOCfish5K-D val (500) | MAE: 15.78 -> 15.19 | Improved |
| increase L_m from 2 to 3 (ablation) | IOCFormer-D · IOCfish5K-D val (500) | MAE: 15.19 -> 15.43 | Hurt |
| add depth modality via cross-agent multimodal fusion | IOCFormer-D · IOCfish5K-D test (2,000) | MAE: 17.12 -> 16.80 | Improved |
| use IOCFormer instead of best crowd counting baseline CLTR | IOCFormer · IOCfish5K test (2,000) | MAE: 18.07 -> 17.12 | Improved |
| use crowd counting methods tuned for DOC (MAN) on indiscernible scenes | MAN · IOCfish5K test (2,000) | MAE: 25.82 | Hurt |
Closest papers by meaning
- NAUTILUS: A Large Multimodal Model for Underwater Scene Understanding (NeurIPS 2025 · similarity 0.84)
- A Unified Image-Dense Annotation Generation Model for Underwater Scenes (CVPR 2025 · similarity 0.84)
- WebUOT-1M: Advancing Deep Underwater Object Tracking with A Million-Scale Benchmark (NeurIPS 2024 · similarity 0.84)
- OmniCount: Multi-label Object Counting with Semantic-Geometric Priors (AAAI 2025 · similarity 0.84)
- CountGD: Multi-Modal Open-World Counting (NeurIPS 2024 · similarity 0.83)
- Diving into Underwater: Segment Anything Model Guided Underwater Salient Instance Segmentation and A Large-scale Dataset (ICML 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.
