To Adapt or Not to Adapt? Real-Time Adaptation for Semantic Segmentation
This enables practical deployment of semantic segmentation systems that can adapt to unforeseen domain changes like weather events, though it is incremental in optimizing existing adaptation methods.
The paper tackles the problem of high computational costs in online domain adaptation for semantic segmentation, proposing HAMLET to achieve real-time adaptation at over 29FPS on a consumer GPU while maintaining accuracy.
The goal of Online Domain Adaptation for semantic segmentation is to handle unforeseeable domain changes that occur during deployment, like sudden weather events. However, the high computational costs associated with brute-force adaptation make this paradigm unfeasible for real-world applications. In this paper we propose HAMLET, a Hardware-Aware Modular Least Expensive Training framework for real-time domain adaptation. Our approach includes a hardware-aware back-propagation orchestration agent (HAMT) and a dedicated domain-shift detector that enables active control over when and how the model is adapted (LT). Thanks to these advancements, our approach is capable of performing semantic segmentation while simultaneously adapting at more than 29FPS on a single consumer-grade GPU. Our framework's encouraging accuracy and speed trade-off is demonstrated on OnDA and SHIFT benchmarks through experimental results.