Efficient Test-time Adaptive Object Detection via Sensitivity-Guided Pruning

2025-06-03

AI Summary

To address the challenges of strong dynamic domain shifts, high computational overhead, and performance degradation caused by sensitive feature interference in Continual Test-Time Adaptation for Object Detection (CTTA-OD), this paper proposes a sensitivity-guided lightweight online adaptation framework. Methodologically, we introduce the first dual-granularity (image- and instance-level) sensitivity quantification mechanism, integrated with weighted sparse regularization and channel-wise pruning, and further propose a stochastic channel reactivation strategy to enable dynamic model slimming and robust parameter updates. Evaluated on three standard benchmarks, our approach achieves state-of-the-art detection accuracy while reducing FLOPs by 12%, significantly improving real-time inference capability and generalization stability under resource-constrained settings. The core contribution lies in the first synergistic integration of sensitivity modeling and stochastic reactivation for CTTA-OD—effectively balancing efficiency, robustness, and adaptability.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Continual test-time adaptive object detection (CTTA-OD) aims to online adapt a source pre-trained detector to ever-changing environments during inference under continuous domain shifts. Most existing CTTA-OD methods prioritize effectiveness while overlooking computational efficiency, which is crucial for resource-constrained scenarios. In this paper, we propose an efficient CTTA-OD method via pruning. Our motivation stems from the observation that not all learned source features are beneficial; certain domain-sensitive feature channels can adversely affect target domain performance. Inspired by this, we introduce a sensitivity-guided channel pruning strategy that quantifies each channel based on its sensitivity to domain discrepancies at both image and instance levels. We apply weighted sparsity regularization to selectively suppress and prune these sensitive channels, focusing adaptation efforts on invariant ones. Additionally, we introduce a stochastic channel reactivation mechanism to restore pruned channels, enabling recovery of potentially useful features and mitigating the risks of early pruning. Extensive experiments on three benchmarks show that our method achieves superior adaptation performance while reducing computational overhead by 12% in FLOPs compared to the recent SOTA method.

Problem

• Adapt object detection to changing domains efficiently
• Reduce computational overhead in continual adaptation
• Balance effectiveness and efficiency in domain shifts

Innovation

• Sensitivity-guided pruning for domain adaptation
• Weighted sparsity regularization for channel suppression
• Stochastic reactivation to recover pruned features