Generalization Under Scrutiny: Cross-Domain Detection Progresses, Pitfalls, and Persistent Challenges

2026-04-09

AI Summary

This work addresses the significant performance degradation of object detection models when deployed across domains due to distributional shifts. It systematically analyzes the unique challenges of cross-domain object detection (CDOD), presenting the first unified framework that elucidates its structural complexities relative to domain adaptation in classification tasks. By modeling the problem through multiple stages, introducing a conceptual taxonomy grounded in adaptation paradigms, modeling assumptions, and detector components, and analyzing how domain shift propagates through the detection pipeline, the study reveals fundamental difficulties inherent to CDOD. Employing a systematic review methodology, it synthesizes existing approaches, datasets, and evaluation protocols, identifies critical bottlenecks, and outlines promising directions for developing robust cross-domain detection systems.

Abstract

Object detection models trained on a source domain often exhibit significant performance degradation when deployed in unseen target domains, due to various kinds of variations, such as sensing conditions, environments and data distributions. Hence, regardless the recent breakthrough advances in deep learning-based detection technology, cross-domain object detection (CDOD) remains a critical research area. Moreover, the existing literature remains fragmented, lacking a unified perspective on the structural challenges underlying domain shift and the effectiveness of adaptation strategies. This survey provides a comprehensive and systematic analysis of CDOD. We start upon a problem formulation that highlights the multi-stage nature of object detection under domain shift. Then, we organize the existing methods through a conceptual taxonomy that categorizes approaches based on adaptation paradigms, modeling assumptions, and pipeline components. Furthermore, we analyze how domain shift propagates across detection stages and discuss why adaptation in object detection is inherently more complex than in classification. In addition, we review commonly used datasets, evaluation protocols, and benchmarking practices. Finally, we identify the key challenges and outline promising future research directions. Cohesively, this survey aims to provide a unified framework for understanding CDOD and to guide the development of more robust detection systems.

Problem

• cross-domain object detection
• domain shift
• generalization
• object detection
• domain adaptation

Innovation

• cross-domain object detection
• domain shift
• adaptation taxonomy
• multi-stage detection
• benchmarking