Global Features are All You Need for Image Retrieval and Reranking

TL;DR

SuperGlobal is a novel image retrieval approach that uses only global features for both retrieval and reranking, achieving high accuracy and efficiency with significant speedup and improved benchmarks.

Contribution

The paper introduces SuperGlobal, a global-feature-only method for image retrieval and reranking, with new modules for GeM pooling and an efficient reranking process, surpassing state-of-the-art performance.

Findings

7.1% improvement on Revisited Oxford+1M Hard dataset

64,865x speedup with two-stage system

16.3% surpassing current single-stage state-of-the-art

Abstract

Image retrieval systems conventionally use a two-stage paradigm, leveraging global features for initial retrieval and local features for reranking. However, the scalability of this method is often limited due to the significant storage and computation cost incurred by local feature matching in the reranking stage. In this paper, we present SuperGlobal, a novel approach that exclusively employs global features for both stages, improving efficiency without sacrificing accuracy. SuperGlobal introduces key enhancements to the retrieval system, specifically focusing on the global feature extraction and reranking processes. For extraction, we identify sub-optimal performance when the widely-used ArcFace loss and Generalized Mean (GeM) pooling methods are combined and propose several new modules to improve GeM pooling. In the reranking stage, we introduce a novel method to update the global features of the query and top-ranked images by only considering feature refinement with a small set of images, thus being very compute and memory efficient. Our experiments demonstrate substantial improvements compared to the state of the art in standard benchmarks. Notably, on the Revisited Oxford+1M Hard dataset, our single-stage results improve by 7.1%, while our two-stage gain reaches 3.7% with a strong 64,865x speedup. Our two-stage system surpasses the current single-stage state-of-the-art by 16.3%, offering a scalable, accurate alternative for high-performing image retrieval systems with minimal time overhead. Code: https://github.com/ShihaoShao-GH/SuperGlobal.