DA-Mamba: Domain Adaptive Hybrid Mamba-Transformer Based One-Stage Object Detection
A. Enes Doruk, Hasan F. Ates
TL;DR
DA-Mamba introduces a domain-adaptive Mamba-Transformer framework for one-stage object detection that blends efficient state-space modeling with attention to capture long-range dependencies across domains. It employs domain-adaptive spatial/channel scanning, cross-attention for source/target dominance, and entropy-guided distillation and perturbation to align domain representations while suppressing noisy activations. The approach demonstrates strong cross-domain performance on Pascal VOC and synthetic cross-domain datasets, achieving state-of-the-art results with favorable computational efficiency. This work advances practical unsupervised domain adaptation for real-world detectors by balancing modeling power with deployment constraints.
Abstract
Recent 2D CNN-based domain adaptation approaches struggle with long-range dependencies due to limited receptive fields, making it difficult to adapt to target domains with significant spatial distribution changes. While transformer-based domain adaptation methods better capture distant relationships through self-attention mechanisms that facilitate more effective cross-domain feature alignment, their quadratic computational complexity makes practical deployment challenging for object detection tasks across diverse domains. Inspired by the global modeling and linear computation complexity of the Mamba architecture, we present the first domain-adaptive Mamba-based one-stage object detection model, termed DA-Mamba. Specifically, we combine Mamba's efficient state-space modeling with attention mechanisms to address domain-specific spatial and channel-wise variations. Our design leverages domain-adaptive spatial and channel-wise scanning within the Mamba block to extract highly transferable representations for efficient sequential processing, while cross-attention modules generate long-range, mixed-domain spatial features to enable robust soft alignment across domains. Besides, motivated by the observation that hybrid architectures introduce feature noise in domain adaptation tasks, we propose an entropy-based knowledge distillation framework with margin ReLU, which adaptively refines multi-level representations by suppressing irrelevant activations and aligning uncertainty across source and target domains. Finally, to prevent overfitting caused by the mixed-up features generated through cross-attention mechanisms, we propose entropy-driven gating attention with random perturbations that simultaneously refine target features and enhance model generalization.