Correlation Verification for Image Retrieval
This work addresses the challenge of geometric verification for image retrieval, offering a novel re-ranking method that significantly boosts accuracy in domain-specific applications.
The paper tackles the re-ranking problem in image retrieval by proposing Correlation Verification Networks (CVNet), which achieved state-of-the-art performance with a 12.6% mAP improvement on the ROxford-Hard+1M benchmark.
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.