# Learning Regional Attention over Multi-resolution Deep Convolutional Features for Trademark Retrieval

arXiv: https://arxiv.org/abs/2104.07240 · HTML: https://mlsanity.com/paper/2104.07240/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 24 references that are in the map.
- First posted on arXiv 15 Apr 2021.

## Builds on (most cited first, 11 of 11)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [Particular object retrieval with integral max-pooling of CNN activations](https://mlsanity.com/paper/1511.05879/) (2015)
- [End-to-end Learning of Deep Visual Representations for Image Retrieval](https://mlsanity.com/paper/1610.07940/) (2016)
- [SIFT Meets CNN: A Decade Survey of Instance Retrieval](https://mlsanity.com/paper/1608.01807/) (2016)
- [Aggregating Deep Convolutional Features for Image Retrieval](https://mlsanity.com/paper/1510.07493/) (2015)
- [Cross-dimensional Weighting for Aggregated Deep Convolutional Features](https://mlsanity.com/paper/1512.04065/) (2015)
- [Class-Weighted Convolutional Features for Visual Instance Search](https://mlsanity.com/paper/1707.02581/) (2017)
- [Component-based Attention for Large-scale Trademark Retrieval](https://mlsanity.com/paper/1811.02746/) (2018)
- [A Large-scale Dataset and Benchmark for Similar Trademark Retrieval](https://mlsanity.com/paper/1701.05766/) (2017)

## Built on by (most cited first, 1 of 1)

- [Segment Augmentation and Differentiable Ranking for Logo Retrieval](https://mlsanity.com/paper/2209.02482/) (2022)

## Abstract (arXiv)

Large-scale trademark retrieval is an important content-based image retrieval task. A recent study shows that off-the-shelf deep features aggregated with Regional-Maximum Activation of Convolutions (R-MAC) achieve state-of-the-art results. However, R-MAC suffers in the presence of background clutter/trivial regions and scale variance, and discards important spatial information. We introduce three simple but effective modifications to R-MAC to overcome these drawbacks. First, we propose the use of both sum and max pooling to minimise the loss of spatial information. We also employ domain-specific unsupervised soft-attention to eliminate background clutter and unimportant regions. Finally, we add multi-resolution inputs to enhance the scale-invariance of R-MAC. We evaluate these three modifications on the million-scale METU dataset. Our results show that all modifications bring non-trivial improvements, and surpass previous state-of-the-art results.