We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CV

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Vision and Pattern Recognition

Title: GPR1200: A Benchmark for General-Purpose Content-Based Image Retrieval

Abstract: Even though it has extensively been shown that retrieval specific training of deep neural networks is beneficial for nearest neighbor image search quality, most of these models are trained and tested in the domain of landmarks images. However, some applications use images from various other domains and therefore need a network with good generalization properties - a general-purpose CBIR model. To the best of our knowledge, no testing protocol has so far been introduced to benchmark models with respect to general image retrieval quality. After analyzing popular image retrieval test sets we decided to manually curate GPR1200, an easy to use and accessible but challenging benchmark dataset with a broad range of image categories. This benchmark is subsequently used to evaluate various pretrained models of different architectures on their generalization qualities. We show that large-scale pretraining significantly improves retrieval performance and present experiments on how to further increase these properties by appropriate fine-tuning. With these promising results, we hope to increase interest in the research topic of general-purpose CBIR.
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
Cite as: arXiv:2111.13122 [cs.CV]
  (or arXiv:2111.13122v1 [cs.CV] for this version)

Submission history

From: Konstantin Schall [view email]
[v1] Thu, 25 Nov 2021 15:19:21 GMT (3560kb,D)

Link back to: arXiv, form interface, contact.