用光照高斯混合模型提升水下显著实例分割精度
USIS-PGM: Photometric Gaussian Mixtures for Underwater Salient Instance Segmentation
- 通过频域感知与动态重加权增强边界特征
- 多尺度高斯热图监督使掩码结构更连贯
- 适合水下机器人视觉理解场景
水下显著实例分割（USIS）对海洋机器人系统至关重要，可实现水下显著目标检测与实例级掩码预测。由于水下图像退化，其挑战性高于陆地场景。本文提出单阶段框架USIS-PGM：编码器通过频域感知模块强化边界特征，并利用动态加权模块进行内容自适应特征重加权；解码器引入基于Transformer的实例激活模块以更好区分显著实例。此外，USIS-PGM采用从真实掩码生成的多尺度光照高斯混合（PGM）热图，对中间解码特征进行监督，从而提升显著实例定位能力并生成结构更一致的掩码。实验表明该模型在性能和实用性上均具优势。
原文摘要 · Abstract (English)
Underwater salient instance segmentation (USIS) is crucial for marine robotic systems, as it enables both underwater salient object detection and instance-level mask prediction for visual scene understanding. Compared with its terrestrial counterpart, USIS is more challenging due to the underwater image degradation. To address this issue, this paper proposes USIS-PGM, a single-stage framework for USIS. Specifically, the encoder enhances boundary cues through a frequency-aware module and performs content-adaptive feature reweighting via a dynamic weighting module. The decoder incorporates a Transformer-based instance activation module to better distinguish salient instances. In addition, USIS-PGM employs multi-scale Gaussian heatmaps generated from ground-truth masks through Photometric Gaussian Mixture (PGM) to supervise intermediate decoder features, thereby improving salient instance localization and producing more structurally coherent mask predictions. Experimental results demonstrate the superiority and practical applicability of the proposed USIS-PGM model.
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。