Faster R-CNN Features for Instance Search

TL;DR

This paper investigates the use of Faster R-CNN features for instance search, leveraging object proposals and CNN features to improve retrieval performance on standard datasets.

Contribution

It introduces a novel instance search pipeline using Faster R-CNN features and evaluates their effectiveness, including fine-tuning for specific objects.

Findings

Faster R-CNN features improve instance retrieval accuracy.

Fine-tuning enhances the discriminative power of features.

Achieves competitive results on benchmark datasets.

Abstract

Image representations derived from pre-trained Convolutional Neural Networks (CNNs) have become the new state of the art in computer vision tasks such as instance retrieval. This work explores the suitability for instance retrieval of image- and region-wise representations pooled from an object detection CNN such as Faster R-CNN. We take advantage of the object proposals learned by a Region Proposal Network (RPN) and their associated CNN features to build an instance search pipeline composed of a first filtering stage followed by a spatial reranking. We further investigate the suitability of Faster R-CNN features when the network is fine-tuned for the same objects one wants to retrieve. We assess the performance of our proposed system with the Oxford Buildings 5k, Paris Buildings 6k and a subset of TRECVid Instance Search 2013, achieving competitive results.