Research map / Detection and segmentation / CNN features
Dynamic Dictionary Learning for Remote Sensing Image Segmentation
Xuechao Zou, Yue Li, Shun Zhang, Kai Li, Shiying Wang, Pin Tao, Junliang Xing, Congyan Lang
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 dynamic loss branch to static loss (ablation) | Dynamic Dictionary Learning · LoveDA | mIoU: 54.80 -> 54.93 | Improved |
| add dynamic loss branch to static loss (ablation) | Dynamic Dictionary Learning · Grass | mIoU: 50.96 -> 51.38 | Improved |
| add contrastive loss to dual-branch losses (ablation) | Dynamic Dictionary Learning · LoveDA | mIoU: 54.95 -> 55.27 | Improved |
| add contrastive loss to dual-branch losses (ablation) | Dynamic Dictionary Learning · Grass | mIoU: 51.53 -> 51.96 | Improved |
| add modulator (ablation) | Dynamic Dictionary Learning · Vaihingen | mIoU: 85.17 -> 85.27 | No clear effect |
| add aggregator (ablation) | Dynamic Dictionary Learning · Vaihingen | mIoU: 82.54 -> 85.27 | Improved |
| add aggregator (ablation) | Dynamic Dictionary Learning · Grass | mIoU: 48.52 -> 51.96 | Improved |
| add interaction module (ablation) | Dynamic Dictionary Learning · LoveDA | mIoU: 54.55 -> 55.27 | Improved |
| use 3 instead of 4 interaction stages (ablation) | Dynamic Dictionary Learning · UAVid | mIoU: 67.88 -> 70.90 | Improved |
| use 3 instead of 4 interaction stages (ablation) | Dynamic Dictionary Learning · Cloud | mIoU: 80.97 -> 83.10 | Improved |
| use 2 instead of 3 interaction stages (ablation) | Dynamic Dictionary Learning · Vaihingen | mF1: 91.76 -> 91.88 | No clear effect |
| fully train ConvNeXt (B) backbone instead of freezing (ablation) | Dynamic Dictionary Learning · LoveDA | mIoU: 48.50 -> 55.27 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SegFormer — SegFormer: A Topic Segmentation Model with Controllable Range of Attention (AAAI 2023) · better on 2 of 2 shared results
- ABCNet — ABCNet: Real-Time Scene Text Spotting With Adaptive Bezier-Curve Network (CVPR 2020) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Sparse and Complete Latent Organization for Geospatial Semantic Segmentation (CVPR 2022 · similarity 0.89)
- Towards Open-Vocabulary Remote Sensing Image Semantic Segmentation (AAAI 2025 · similarity 0.88)
- ReAttnCLIP: Training-Free Open-Vocabulary Remote Sensing Image Segmentation via Re-defined Attention in CLIP (CVPR 2026 · similarity 0.88)
- InstructSAM: A Training-free Framework for Instruction-Oriented Remote Sensing Object Recognition (NeurIPS 2025 · similarity 0.87)
- ZoRI: Towards Discriminative Zero-Shot Remote Sensing Instance Segmentation (AAAI 2025 · similarity 0.86)
- DiffCLIP: Few-shot Language-driven Multimodal Classifier (AAAI 2025 · 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.
