Research map / Detection and segmentation / Open-set and few-shot detection
Improving Target Presence and Plurality Recognition for Generalized Referring Image Segmentation
Namyup Kim, Jinsung Lee, Suha Kwak
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 target query for presence classification (ablation) | generalized RIS model · gRefCOCO val | gIoU: 62.39 -> 64.93 | Improved |
| add target query for presence classification (ablation) | generalized RIS model · gRefCOCO val | cIoU: 59.62 -> 63.86 | Improved |
| add plurality classification head on target query (ablation) | generalized RIS model with target query · gRefCOCO val | gIoU: 64.93 -> 66.11 | Improved |
| add no-target data augmentation (ablation) | target query plus plurality model · gRefCOCO val | N-acc: 57.42 -> 71.28 | Improved |
| add no-target data augmentation (ablation) | target query plus plurality model · gRefCOCO val | cIoU: 65.05 -> 63.61 | Hurt |
| add multi-target data augmentation on top of no-target augmentation (ablation) | target query plus plurality model · gRefCOCO val | cIoU: 63.61 -> 66.61 | Improved |
| add multi-target data augmentation on top of no-target augmentation (ablation) | target query plurality model · gRefCOCO val | gIoU: 69.24 -> 71.32 | Improved |
| use proposed architecture instead of ReLA | generalized RIS · gRefCOCO val | N-acc: 56.37 -> 72.27 | Improved |
| use proposed model instead of GSVA (LMM-based) | generalized RIS · gRefCOCO val | gIoU: 66.47 -> 71.32 | Improved |
| use proposed model instead of GSVA on testB | generalized RIS · gRefCOCO testB | gIoU: 62.23 -> 62.24 | No clear effect |
| attach ReLA's presence head (ablation) | ReLA · gRefCOCO val | gIoU: +0.27%p | No clear effect |
| replace ReLA presence prediction with ground-truth presence labels (ablation) | ReLA · gRefCOCO val | gIoU: +18.04%p | Improved |
Closest papers by meaning
- LQMFormer: Language-aware Query Mask Transformer for Referring Image Segmentation (CVPR 2024 · similarity 0.90)
- Mask Grounding for Referring Image Segmentation (CVPR 2024 · similarity 0.89)
- Hybrid Global-Local Representation with Augmented Spatial Guidance for Zero-Shot Referring Image Segmentation (CVPR 2025 · similarity 0.89)
- Prompt-Driven Referring Image Segmentation with Instance Contrasting (CVPR 2024 · similarity 0.89)
- GSVA: Generalized Segmentation via Multimodal Large Language Models (CVPR 2024 · similarity 0.89)
- IteRPrimE: Zero-shot Referring Image Segmentation with Iterative Grad-CAM Refinement and Primary Word Emphasis (AAAI 2025 · similarity 0.87)
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.
