Hihpq: Hierarchical Hyperbolic Product Quantization For Unsupervised Image Retrieval

Abstract

Existing unsupervised deep product quantization methods primarily aim for the increased similarity between different views of the identical image, whereas the delicate multi-level semantic similarities preserved between images are overlooked. Moreover, these methods predominantly focus on the Euclidean space for computational convenience, compromising their ability to map the multi-level semantic relationships between images effectively. To mitigate these shortcomings, we propose a novel unsupervised product quantization method dubbed \textbf\{Hi\}erarchical \textbf\{H\}yperbolic \textbf\{P\}roduct \textbf\{Q\}uantization (HiHPQ), which learns quantized representations by incorporating hierarchical semantic similarity within hyperbolic geometry. Specifically, we propose a hyperbolic product quantizer, where the hyperbolic codebook attention mechanism and the quantized contrastive learning on the hyperbolic product manifold are introduced to expedite quantization. Furthermore, we propos

Stats

• citations7
• S2 citations—
• github stars0
• HF likes0
• heat score6.77
• arxiv keyqiu2024hihpq

Related papers

• End-to-end Supervised Product Quantization For Image Search And Retrieval (2017)13.05
• Generalized Product Quantization Network For Semi-supervised Image Retrieval (2020)12.17
• Self-supervised Product Quantization For Deep Unsupervised Image Retrieval (2021)13.44
• Beyond Product Quantization: Deep Progressive Quantization For Image Retrieval (2019)12.95
• Weakly Supervised Deep Hyperspherical Quantization For Image Retrieval (2024)10.04
• Matching-oriented Product Quantization For Ad-hoc Retrieval (2021)2.29
• PHPQ: Pyramid Hybrid Pooling Quantization For Efficient Fine-grained Image Retrieval (2021)8.60
• Orthonormal Product Quantization Network For Scalable Face Image Retrieval (2021)8.60