Research map / Detection and segmentation / Open-set and few-shot detection
Multimodal Referring Segmentation: A Survey
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 |
|---|---|---|---|
| modern fully-supervised transformer vs early LSTM-CNN | OneRef-L · RefCOCOg val (UMD) | mIoU: 34.06 -> 75.68 | Improved |
| MLLM-based SAM4MLLM vs conventional OneRef-L on test | SAM4MLLM-8B · RefCOCOg test (UMD) | mIoU: 76.82 -> 76.40 | No clear effect |
| weak/semi-supervision (ADDP) vs full supervision | ADDP · RefCOCOg val (UMD) | mIoU: 59.05 -> 75.68 | Mixed |
| recent RVOS method vs URVOS | VRS-HQ · Ref-YouTube-VOS | J&F: 47.23 -> 71.00 | Improved |
| GLUS vs VRS-HQ on MeViS | GLUS · MeViS | J&F: 50.90 -> 51.30 | No clear effect |
| recent AVS method vs original AVSBench baseline | DDESeg · AVSBench S4 | J: 78.74 -> 92.40 | Improved |
| DDESeg on harder AVSS benchmark | DDESeg · AVSBench-Semantic AVSS | J: 29.77 -> 63.40 | Improved |
| weakly-supervised WS-AVS vs fully-supervised RAVS | WS-AVS · AVSBench S4 | J: 34.13 -> 93.10 | Hurt |
| OISA-1B vs LISA-13B on OmniAVS | OISA-1B · OmniAVS (All) | J&F: 36.1 -> 41.1 | Improved |
| larger LISA model | LISA · OmniAVS (All) | J&F: 33.6 -> 36.1 | Improved |
| OISA-1B on Ref-AVS seen vs unseen subsets | OISA-1B · Ref-AVS | J: 51.7 -> 58.3 | Mixed |
Closest papers by meaning
- Recurrent Multimodal Interaction for Referring Image Segmentation (ICCV 2017 · similarity 0.88)
- Locate Then Segment: A Strong Pipeline for Referring Image Segmentation (CVPR 2021 · similarity 0.86)
- Referring Image Segmentation via Cross-Modal Progressive Comprehension (CVPR 2020 · similarity 0.86)
- Prompt-Driven Referring Image Segmentation with Instance Contrasting (CVPR 2024 · similarity 0.85)
- GSVA: Generalized Segmentation via Multimodal Large Language Models (CVPR 2024 · similarity 0.84)
- SetCon: Towards Open-Ended Referring Segmentation via Set-Level Concept Prediction (NeurIPS 2026 · 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.
