Link and code: Fast indexing with graphs and compact regression codes
This addresses the memory and speed challenges for indexing billions of images on a single server, representing an incremental improvement over existing graph-based methods.
The paper tackles the problem of efficient similarity search for billion-scale image indexing under memory constraints by combining graph traversal with compact quantization codes, achieving state-of-the-art performance with 64-128 bytes per vector on public benchmarks.
Similarity search approaches based on graph walks have recently attained outstanding speed-accuracy trade-offs, taking aside the memory requirements. In this paper, we revisit these approaches by considering, additionally, the memory constraint required to index billions of images on a single server. This leads us to propose a method based both on graph traversal and compact representations. We encode the indexed vectors using quantization and exploit the graph structure to refine the similarity estimation. In essence, our method takes the best of these two worlds: the search strategy is based on nested graphs, thereby providing high precision with a relatively small set of comparisons. At the same time it offers a significant memory compression. As a result, our approach outperforms the state of the art on operating points considering 64-128 bytes per vector, as demonstrated by our results on two billion-scale public benchmarks.