论文标题
图像检索的相关验证
Correlation Verification for Image Retrieval
论文作者
论文摘要
几何验证被认为是图像检索中重新排列任务的事实上的解决方案。在这项研究中，我们提出了一个新的图像检索重新排列网络，名为相关验证网络（CVNET）。我们提出的网络包括深堆积的4D卷积层，逐渐将密集的特征相关性压缩到图像相似性中，同时从各种图像对学习各种几何匹配模式。为了启用跨尺度匹配，它构建了特征金字塔并在单个推理中构造跨尺度特征相关性，以取代昂贵的多尺度推断。此外，我们将课程学习与硬采矿和捉迷藏的策略一起使用，以处理硬样品而不会失去一般性。我们提出的重新排列网络在几个检索基准测试基准上显示了最先进的性能，并在最先进的方法上具有明显的利润率（Roxford-Hard-Hard-Hard+1m集的地图+12.6％）。源代码和模型可在线获得：https：//github.com/sungonce/cvnet。
Geometric verification is considered a de facto solution for the re-ranking task in image retrieval. In this study, we propose a novel image retrieval re-ranking network named Correlation Verification Networks (CVNet). Our proposed network, comprising deeply stacked 4D convolutional layers, gradually compresses dense feature correlation into image similarity while learning diverse geometric matching patterns from various image pairs. To enable cross-scale matching, it builds feature pyramids and constructs cross-scale feature correlations within a single inference, replacing costly multi-scale inferences. In addition, we use curriculum learning with the hard negative mining and Hide-and-Seek strategy to handle hard samples without losing generality. Our proposed re-ranking network shows state-of-the-art performance on several retrieval benchmarks with a significant margin (+12.6% in mAP on ROxford-Hard+1M set) over state-of-the-art methods. The source code and models are available online: https://github.com/sungonce/CVNet.
