# One-Shot Texture Retrieval with Global Context Metric

## TL;DR

This paper introduces a novel one-shot texture retrieval method that uses a directionality-aware module and global context to improve texture segmentation accuracy and generalization from minimal examples.

## Contribution

The paper presents a new OS-TR network with a directionality-aware module and self-gating mechanism for effective one-shot texture retrieval and segmentation.

## Key findings

- Achieves superior segmentation performance on benchmark datasets.
- Demonstrates robust generalization across different domains.
- Outperforms existing methods in one-shot texture retrieval tasks.

## Abstract

In this paper, we tackle one-shot texture retrieval: given an example of a new reference texture, detect and segment all the pixels of the same texture category within an arbitrary image. To address this problem, we present an OS-TR network to encode both reference and query image, leading to achieve texture segmentation towards the reference category. Unlike the existing texture encoding methods that integrate CNN with orderless pooling, we propose a directionality-aware module to capture the texture variations at each direction, resulting in spatially invariant representation. To segment new categories given only few examples, we incorporate a self-gating mechanism into relation network to exploit global context information for adjusting per-channel modulation weights of local relation features. Extensive experiments on benchmark texture datasets and real scenarios demonstrate the above-par segmentation performance and robust generalization across domains of our proposed method.