论文标题
通过光谱群集投票无监督的显着对象检测
Unsupervised Salient Object Detection with Spectral Cluster Voting
论文作者
论文摘要
在本文中，我们通过利用频谱聚类在自我监督的特征上，解决了无监督的显着对象检测（SOD）的挑战性任务。我们做出以下贡献：（i）我们重新访问频谱聚类并证明其潜力将显着物体的像素分组； （ii）给定来自多个应用程序的掩蔽提案，这些谱集聚类是根据各种自我监督模型计算出的图像特征的，例如Mocov2，Swav，Dino，我们提出了一种简单但有效的胜利者 - 全部投票机制，用于选择明显的面具，利用基于帧和独特性的利用对象授权； （iii）使用选定的对象分割作为伪地面掩模，我们训练一个名为“自mask”的显着对象检测器，该对象检测器的表现优于先前的三个无监督的SOD基准。代码可在https://github.com/noelshin/selfmask上公开获取。
In this paper, we tackle the challenging task of unsupervised salient object detection (SOD) by leveraging spectral clustering on self-supervised features. We make the following contributions: (i) We revisit spectral clustering and demonstrate its potential to group the pixels of salient objects; (ii) Given mask proposals from multiple applications of spectral clustering on image features computed from various self-supervised models, e.g., MoCov2, SwAV, DINO, we propose a simple but effective winner-takes-all voting mechanism for selecting the salient masks, leveraging object priors based on framing and distinctiveness; (iii) Using the selected object segmentation as pseudo groundtruth masks, we train a salient object detector, dubbed SelfMask, which outperforms prior approaches on three unsupervised SOD benchmarks. Code is publicly available at https://github.com/NoelShin/selfmask.
