Research map / Detection and segmentation / Open-set and few-shot detection
RIS-LAD: A Benchmark and Model for Referring Image Segmentation in Low-Altitude Drone Imagery
Kai Ye, YingShi Luan, Zhudi Chen, Guangyue Meng, Pingyang Dai, Liujuan Cao
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 CDLE (class-level linguistic injection in encoder) (ablation) | SAARN · RIS-LAD val | mIoU: 43.97 -> 44.30 | Improved |
| add CDLE (ablation) | SAARN · RIS-LAD test | mIoU: 40.68 -> 41.67 | Improved |
| add ARFM (adaptive reasoning fusion) (ablation) | SAARN · RIS-LAD val | oIoU: 49.82 -> 51.54 | Improved |
| add ARFM (ablation) | SAARN with CDLE · RIS-LAD test | oIoU: 49.28 -> 49.60 | Mixed |
| add CDLE (ablation) | SAARN with ARFM · RIS-LAD test | oIoU: 49.70 -> 49.60 | No clear effect |
| add global linguistic feature l to ARFM (ablation) | ARFM · RIS-LAD val | oIoU: 49.82 -> 50.64 | Improved |
| add descriptive feature d alongside l without class feature c (ablation) | ARFM · RIS-LAD val | oIoU: 50.64 -> 49.90 | Hurt |
| add class-level feature c to ARFM (ablation) | ARFM · RIS-LAD val | oIoU: 50.64 -> 51.06 | Improved |
| add descriptive feature d on top of l and c (ablation) | ARFM · RIS-LAD val | oIoU: 51.06 -> 51.54 | Improved |
| use frozen CLIP image encoder for visual enhancement | VATEX · RIS-LAD val | oIoU: 24.83 (vs LAVT 44.03) | Hurt |
| use CLIP and SAM pretrained models | RSRefSeg · RIS-LAD val | P@0.9: P@0.9 26.58 vs SAARN 19.27; oIoU 50.04 vs 51.54 | Mixed |
Closest papers by meaning
- PixDLM: A Dual-Path Multimodal Language Model for UAV Reasoning Segmentation (CVPR 2026 · similarity 0.89)
- Unleashing Multispectral Video's Potential in Semantic Segmentation: A Semi-supervised Viewpoint and New UAV-View Benchmark (NeurIPS 2024 · similarity 0.86)
- High Temporal Consistency through Semantic Similarity Propagation in Semi-Supervised Video Semantic Segmentation for Autonomous Flight (CVPR 2025 · similarity 0.86)
- Aerial Lifting: Neural Urban Semantic and Building Instance Lifting from Aerial Imagery (CVPR 2024 · similarity 0.86)
- Mask Grounding for Referring Image Segmentation (CVPR 2024 · similarity 0.85)
- WeatherPrompt: Multi-modality Representation Learning for All-Weather Drone Visual Geo-Localization (NeurIPS 2025 · similarity 0.84)
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.
