Segment Augmentation and Differentiable Ranking for Logo Retrieval

TL;DR

This paper introduces a segment-based augmentation method and evaluates a ranking loss function to improve deep logo retrieval, demonstrating enhanced performance on a large dataset.

Contribution

It proposes a novel segment augmentation strategy and assesses Smooth-AP loss, showing their effectiveness for logo retrieval tasks.

Findings

Segment augmentation improves retrieval accuracy.

Smooth-AP outperforms traditional loss functions.

Method achieves better results on METU Trademark Dataset.

Abstract

Logo retrieval is a challenging problem since the definition of similarity is more subjective compared to image retrieval tasks and the set of known similarities is very scarce. To tackle this challenge, in this paper, we propose a simple but effective segment-based augmentation strategy to introduce artificially similar logos for training deep networks for logo retrieval. In this novel augmentation strategy, we first find segments in a logo and apply transformations such as rotation, scaling, and color change, on the segments, unlike the conventional image-level augmentation strategies. Moreover, we evaluate whether the recently introduced ranking-based loss function, Smooth-AP, is a better approach for learning similarity for logo retrieval. On the large scale METU Trademark Dataset, we show that (i) our segment-based augmentation strategy improves retrieval performance compared to the baseline model or image-level augmentation strategies, and (ii) Smooth-AP indeed performs better than conventional losses for logo retrieval.