AFRDA：面向領域自適應語義分割的注意力特徵精煉

摘要

在無監督領域自適應語義分割（UDA-SS）中，模型首先在有標籤的源域數據（例如合成圖像）上進行訓練，然後在無標籤的目標域（例如真實世界圖像）上進行適應，而無需訪問目標域的註釋。現有的UDA-SS方法往往難以平衡細粒度的局部細節與全局上下文信息，導致在複雜區域出現分割錯誤。為解決這一問題，我們引入了自適應特徵精煉（AFR）模塊，該模塊通過利用低分辨率logits中的語義先驗來精煉高分辨率特徵，從而提升分割精度。AFR還整合了高頻成分，這些成分捕捉了細粒度結構並提供了關鍵的邊界信息，改善了物體輪廓的描繪。此外，AFR通過不確定性驅動的注意力機制自適應地平衡局部與全局信息，減少了誤分類。其輕量級設計使其能夠無縫集成到基於HRDA的UDA方法中，實現了最先進的分割性能。我們的方法在GTA V --> Cityscapes上將現有UDA-SS方法的mIoU提升了1.05%，在Synthia-->Cityscapes上提升了1.04%。我們框架的實現可在此處獲取：https://github.com/Masrur02/AFRDA。

English

In Unsupervised Domain Adaptive Semantic Segmentation (UDA-SS), a model is trained on labeled source domain data (e.g., synthetic images) and adapted to an unlabeled target domain (e.g., real-world images) without access to target annotations. Existing UDA-SS methods often struggle to balance fine-grained local details with global contextual information, leading to segmentation errors in complex regions. To address this, we introduce the Adaptive Feature Refinement (AFR) module, which enhances segmentation accuracy by refining highresolution features using semantic priors from low-resolution logits. AFR also integrates high-frequency components, which capture fine-grained structures and provide crucial boundary information, improving object delineation. Additionally, AFR adaptively balances local and global information through uncertaintydriven attention, reducing misclassifications. Its lightweight design allows seamless integration into HRDA-based UDA methods, leading to state-of-the-art segmentation performance. Our approach improves existing UDA-SS methods by 1.05% mIoU on GTA V --> Cityscapes and 1.04% mIoU on Synthia-->Cityscapes. The implementation of our framework is available at: https://github.com/Masrur02/AFRDA

AFRDA：面向領域自適應語義分割的注意力特徵精煉

AFRDA: Attentive Feature Refinement for Domain Adaptive Semantic Segmentation

摘要

Support