mlsanityData to 7 Oct 2026

Remedying Target-Domain Astigmatism for Cross-Domain Few-Shot Object Detection

arXiv 2603.18541 ↗First posted 19 Mar 2026cs.CVcited by 2

Not on the map. Why?

Abstract (arXiv)

Cross-domain few-shot object detection (CD-FSOD) aims to adapt pretrained detectors from a source domain to target domains with limited annotations, suffering from severe domain shifts and data scarcity problems. In this work, we find a previously overlooked phenomenon: models exhibit dispersed and unfocused attention in target domains, leading to imprecise localization and redundant predictions, just like a human cannot focus on visual objects. Therefore, we call it the target-domain Astigmatism problem. Analysis on attention distances across transformer layers reveals that regular fine-tuning inherently shows a trend to remedy this problem, but results are still far from satisfactory, which we aim to enhance in this paper. Biologically inspired by the human fovea-style visual system, we enhance the fine-tuning's inherent trend through a center-periphery attention refinement framework, which contains (1) a Positive Pattern Refinement module to reshape attention toward semantic objects using class-specific prototypes, simulating the visual center region; (2) a Negative Context Modulation module to enhance boundary discrimination by modeling background context, simulating the visual periphery region; and (3) a Textual Semantic Alignment module to strengthen center-periphery distinction through cross-modal cues. Our bio-inspired approach transforms astigmatic attention into focused patterns, substantially improving adaptation to target domains. Experiments on six challenging CD-FSOD benchmarks consistently demonstrate improved detection accuracy and establish new state-of-the-art results.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 25 of its 47 references that are in the map. First posted on arXiv 19 Mar 2026.

Builds on

25 of 47 references.

References · 25

  1. 2021Grounded Language-Image Pre-trainingcited by 1,147
  2. 2022Exploring Plain Vision Transformer Backbones for Object Detectioncited by 723
  3. 2016Multi-View 3D Object Detection Network for Autonomous Drivingcited by 721
  4. 2022Detecting Twenty-thousand Classes using Image-level Supervisioncited by 613
  5. 2019Object Detection in Optical Remote Sensing Images: A Survey and A New Benchmarkcited by 339
  6. 2018Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptationcited by 218
  7. 2019Segmentation-Based Deep-Learning Approach for Surface-Defect Detectioncited by 145
  8. 2023What Do Self-Supervised Vision Transformers Learn?cited by 90
  9. 2020A Realistic Fish-Habitat Dataset to Evaluate Algorithms for Underwater Visual Analysiscited by 57
  10. 2022Margin-Based Few-Shot Class-Incremental Learning with Class-Level Overfitting Mitigationcited by 45
  11. 2024Cross-Domain Few-Shot Object Detection via Enhanced Open-Set Object Detectorcited by 43
  12. 2019Few-shot Adaptive Faster R-CNNcited by 42
13 more
  1. 2020Domain-Adaptive Few-Shot Learningcited by 40
  2. 2024Flatten Long-Range Loss Landscapes for Cross-Domain Few-Shot Learningcited by 26
  3. 2022AcroFOD: An Adaptive Method for Cross-domain Few-shot Object Detectioncited by 25
  4. 2024NegativePrompt: Leveraging Psychology for Large Language Models Enhancement via Negative Emotional Stimulicited by 18
  5. 2024Compositional Few-Shot Class-Incremental Learningcited by 18
  6. 2022CD-FSOD: A Benchmark for Cross-domain Few-shot Object Detectioncited by 17
  7. 2025Domain-RAG: Retrieval-Guided Compositional Image Generation for Cross-Domain Few-Shot Object Detectioncited by 14
  8. 2025Revisiting Pool-based Prompt Learning for Few-shot Class-incremental Learningcited by 8
  9. 2024Learning Unknowns from Unknowns: Diversified Negative Prototypes Generator for Few-Shot Open-Set Recognitioncited by 6
  10. 2026Reclaiming Lost Text Layers for Source-Free Cross-Domain Few-Shot Learningcited by 5
  11. 2025Decoupling Template Bias in CLIP: Harnessing Empty Prompts for Enhanced Few-Shot Learningcited by 4
  12. 2026Mind the Discriminability Trap in Source-Free Cross-domain Few-shot Learningcited by 4
  13. 2026Interpretable Cross-Domain Few-Shot Learning with Rectified Target-Domain Local Alignmentcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Free-Lunch Augmentation by Revisiting Diffusion-Based Data Generation for Cross-Domain Few-Shot Object Detectioncited by 0
  2. 2026The First EgoCross Challenge at EgoVis 2026: Cross-Domain Egocentric Video Question Answeringcited by 0