融合通用与任务特有特征，提升玻璃分割精度。
Glass Segmentation with Fusion of Learned and General Visual Features
- 双骨干网络分别提取通用与任务特定视觉特征
- 在四个数据集上达到当前最佳分割性能
- 兼顾精度与推理速度，适合机器人场景
从RGB图像中进行玻璃表面分割是一项挑战性任务，因为玻璃作为透明材料显著缺乏视觉特征。然而，玻璃分割对场景理解与机器人应用至关重要，需将透明玻璃表面识别为实体材料。本文提出一种新型架构，采用双骨干网络生成通用视觉特征与任务特定学习特征。通用特征由冻结的DINOv3视觉基础模型生成，任务特定特征则通过监督训练的Swin模型获取。多尺度特征表示经残差压缩-激励通道缩减后，输入Mask2Former解码器，生成最终分割掩码。该架构在四个常用玻璃分割数据集上评估，多个准确率指标达到当前最优。模型推理速度优于先前最优方法，且使用更轻量级的DINOv3变体时表现更优。代码与模型权重已公开于：https://github.com/ojalar/lgnet
原文摘要 · Abstract (English)
Glass surface segmentation from RGB images is a challenging task, since glass as a transparent material distinctly lacks visual characteristics. However, glass segmentation is critical for scene understanding and robotics, as transparent glass surfaces must be identified as solid material. This paper presents a novel architecture for glass segmentation, deploying a dual-backbone producing general visual features as well as task-specific learned visual features. General visual features are produced by a frozen DINOv3 vision foundation model, and the task-specific features are generated with a Swin model trained in a supervised manner. Resulting multi-scale feature representations are downsampled with residual Squeeze-and-Excitation Channel Reduction, and fed into a Mask2Former Decoder, producing the final segmentation masks. The architecture was evaluated on four commonly used glass segmentation datasets, achieving state-of-the-art results on several accuracy metrics. The model also has a competitive inference speed compared to the previous state-of-the-art method, and surpasses it when using a lighter DINOv3 backbone variant. The implementation source code and model weights are available at: https://github.com/ojalar/lgnet
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。