Lune

CVPR Findings2026

GReD-RSITR: A Generative Re-Examined Discriminative Framework for Remote Sensing Image-Text Retrieval

Shuhuai Wang, Songwei Pei, Bingfeng Liu, Yuanzhou Huang, Qian Li, Shangguang Wang

Abstract

Remote sensing image-text retrieval (RSITR) aims to achieve accurate matching between remote sensing images and textual descriptions. However, existing methods rely on discriminative models, focusing on mining highly discriminative features, while neglecting the modeling of the joint distribution patterns of image-text pairs. In this paper, we propose a novel Generative Re-examined Discriminative framework for RSITR (GReD) to effectively model joint distribution patterns and re-examine the discriminative space. Specifically, we propose a Semantic Enhancement Discriminative Sub-Network (SEDN) to mine more discriminative features by designing multiple RS-adapters. Subsequently, we propose a Target Enhancement Generative Sub-Network (TEGN) to model the joint distribution patterns of imagetext pairs by designing RS-Diffusion. Meanwhile, we propose that TEGN re-examines SEDN to combine the advantages of discriminative and generative models by integrating them into a unified framework. Additionally, we propose a two-stage training strategy to further optimize the entire GReD framework. Extensive experimental results demonstrate that GReD significantly outperforms existing state-of-the-art methods across multiple metrics.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing cites High-Resolution Image Synthesis with Latent Diffusion Models
  • BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing cites BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
  • BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing cites Learning Transferable Visual Models From Natural Language Supervision
  • AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognition cites Learning Transferable Visual Models From Natural Language Supervision
  • BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation cites Learning Transferable Visual Models From Natural Language Supervision
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on8

  1. Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
  2. High-Resolution Image Synthesis with Latent Diffusion ModelsRobin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser et al.CVPR 2022 · 13,123 citations
  3. BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and GenerationJunnan Li, Dongxu Li, Caiming Xiong, Steven C. H. HoiICML 2022 · 6,549 citations
  4. AdaptFormer: Adapting Vision Transformers for Scalable Visual RecognitionShoufa Chen, Chongjian Ge, Zhan Tong, Jiangliu Wang et al.NeurIPS 2022 · 1,291 citations
  5. BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and EditingDongxu Li, Junnan Li, Steven C. H. HoiNeurIPS 2023 · 587 citations

Related papers

  1. Remote Sensing Image Super-Resolution for Imbalanced Textures: A Texture-Aware Diffusion FrameworkEnzhuo Zhang, Sijie Zhao, Dilxat Muhtar, Zhenshi Li et al.CVPR 2026
  2. RS2-SAM2: Customized SAM2 for Referring Remote Sensing Image SegmentationFu Rong, Meng Lan, Qian Zhang, Lefei ZhangAAAI 2026 · 1 citation
  3. Robust Remote Sensing Image–Text Retrieval with Noisy Correspondenceqiya song, Yiqiang Xie, Yuan Sun, Renwei Dian et al.CVPR 2026 · 1 citation
  4. Entity-Level Alignment with Prompt-Guided Adapter for Remote Sensing Image-Text RetrievalShuoshuo Li, Shuli Cheng, Liejun WangACM MM 2025 · 2 citations
  5. Improving Fusion of Region Features and Grid Features via Two-Step Interaction for Image-Text RetrievalDongqing Wu, Huihui Li, Cang Gu, Lei Guo et al.ACM MM 2022 · 10 citations
Dusk over the sea between two cliffs drawn in fine vertical lines