实时自适应分割模型，动态应对测试时域偏移与语义模糊。
DAWA: Dynamic Ambiguity-Wise Adaptation for Real-Time Domain Adaptive Semantic Segmentation
- 根据测试时域变化动态调整策略，避免错误累积。
- 在Cityscapes、CamVid等数据集上达到40 FPS实时速度，性能领先。
- 适合需要快速部署的自动驾驶、机器人视觉场景。
测试时域自适应（TTDA）语义分割旨在将源域训练的分割模型实时适配到目标域进行推理，效率与效果至关重要。然而，现有方法或依赖昂贵的逐帧优化，或假设不现实的域偏移，导致适应效率低且持续存在语义模糊。为此，本文提出一种实时TTDA语义分割框架——动态模糊感知自适应（DAWA），可自适应检测域偏移，并动态调整学习策略以缓解测试时的持续模糊问题。具体提出动态模糊块掩码（DAP Mask）策略，动态识别并掩码高干扰区域，防止模糊类别中的误差累积；同时提出动态模糊类别混合（DAC Mix）策略，利用视觉-语言模型对语义相似类别分组，并通过元模糊类别缓冲区增强目标域数据。在多个主流TTDA基准上的实验表明，DAWA持续优于现有最优方法，且保持约40 FPS的实时推理速度。
原文摘要 · Abstract (English)
Test-time domain adaption (TTDA) for semantic segmentation aims to adapt a segmentation model trained on a source domain to a target domain for inference on-the-fly, where both efficiency and effectiveness are critical. However, existing TTDA methods either rely on costly frame-wise optimization or assume unrealistic domain shifts, resulting in poor adaptation efficiency and continuous semantic ambiguities. To address these challenges, we propose a real-time framework for TTDA semantic segmentation, called Dynamic Ambiguity-Wise Adaptation (DAWA), which adaptively detects domain shifts and dynamically adjusts the learning strategies to mitigate continuous ambiguities in the test time. Specifically, we introduce the Dynamic Ambiguous Patch Mask (DAP Mask) strategy, which dynamically identifies and masks highly disturbed regions to prevent error accumulation in ambiguous classes. Furthermore, we present the Dynamic Ambiguous Class Mix (DAC Mix) strategy that leverages vision-language models to group semantically similar classes and augment the target domain with a meta-ambiguous class buffer. Extensive experiments on widely used TTDA benchmarks demonstrate that DAWA consistently outperforms state-of-the-art methods, while maintaining real-time inference speeds of approximately 40 FPS.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。