Research map / Detection and segmentation / Open-set and few-shot detection
GREx: Generalized Referring Expression Segmentation, Comprehension, and Generation
Henghui Ding, Chang Liu, Shuting He, Xudong Jiang, Yu-Gang Jiang
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 RIA adaptive region split instead of hard split at input (ablation) | ReLA · gRefCOCO val | Pr@F1: 53.26 -> 61.90 | Improved |
| add minimap to RIA (ablation) | ReLA · gRefCOCO val | Pr@F1: 60.07 -> 61.90 | Improved |
| add language attention to RLA baseline (ablation) | ReLA · gRefCOCO val | Pr@F1: 56.03 -> 58.26 | Improved |
| add region attention on top of language attention (ablation) | ReLA · gRefCOCO val | Pr@F1: 58.26 -> 59.86 | Improved |
| use full RLA with region-region and region-language dependencies (ablation) | ReLA · gRefCOCO val | Pr@F1: 59.86 -> 61.90 | Improved |
| increase number of regions from 4x4 to 10x10 (ablation) | ReLA · gRefCOCO val | Pr@F1: 55.18 -> 61.90 | Improved |
| increase regions from 10x10 to 12x12 (ablation) | ReLA · gRefCOCO val | Pr@F1: 61.90 -> 61.04 | Hurt |
| use counting head N_O instead of threshold output (ablation) | ReLA · gRefCOCO val | Pr@F1: 58.24 -> 61.90 | Improved |
| use counting head N_O instead of binary classifier (ablation) | ReLA · gRefCOCO val | Pr@F1: 37.58 -> 61.90 | Improved |
| add multi-task training of GRES and GREC (ablation) | ReLA · gRefCOCO val | Pr@F1: 61.58 -> 61.90 | Improved |
| plug ReLA into LAVT | LAVT · gRefCOCO val | cIoU: 57.64 -> 61.23 | Improved |
| use full ReLA instead of LAVT with ReLA | ReLA · gRefCOCO val | cIoU: 61.23 -> 62.91 | Improved |
Closest papers by meaning
- GSVA: Generalized Segmentation via Multimodal Large Language Models (CVPR 2024 · similarity 0.88)
- Unveiling Parts Beyond Objects: Towards Finer-Granularity Referring Expression Segmentation (CVPR 2024 · similarity 0.87)
- Hierarchical Alignment-enhanced Adaptive Grounding Network for Generalized Referring Expression Comprehension (AAAI 2025 · similarity 0.86)
- RefDetector: A Simple Yet Effective Matching-based Method for Referring Expression Comprehension (AAAI 2025 · similarity 0.86)
- Rethinking Two-Stage Referring Expression Comprehension: A Novel Grounding and Segmentation Method Modulated by Point (AAAI 2024 · similarity 0.86)
- ScanFormer: Referring Expression Comprehension by Iteratively Scanning (CVPR 2024 · similarity 0.85)
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.
