---
title: "Domain-Smoothing Network for Zero-Shot Sketch-Based Image Retrieval"
authors:
  - "Zhipeng Wang"
  - "Hao Wang"
  - "Jiexi Yan"
  - "Aming Wu"
  - "Cheng Deng"
doi: "10.24963/ijcai.2021/158"
url: "https://ijcai.org/proceedings/2021/158"
pdf: "https://ijcai.org/proceedings/2021/0158.pdf"
venue: "Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence"
year: 2021
pages: "1143-1149"
track: "Main Track"
keywords:
  - "Machine Learning: Transfer, Adaptation, Multi-task Learning"
  - "Computer Vision: Recognition: Detection, Categorization, Indexing, Matching, Retrieval, Semantic Interpretation"
---

# Domain-Smoothing Network for Zero-Shot Sketch-Based Image Retrieval

Zhipeng Wang, Hao Wang, Jiexi Yan, Aming Wu, Cheng Deng

*Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence*, pages 1143-1149. DOI: [10.24963/ijcai.2021/158](https://doi.org/10.24963/ijcai.2021/158)

## Abstract

Zero-Shot Sketch-Based Image Retrieval (ZS-SBIR) is a novel cross-modal retrieval task, where abstract sketches are used as queries to retrieve natural images under zero-shot scenario. Most existing methods regard ZS-SBIR as a traditional classification problem and employ a cross-entropy or triplet-based loss to achieve retrieval, which neglect the problems of the domain gap between sketches and natural images and the large intra-class diversity in sketches. Toward this end, we propose a novel Domain-Smoothing Network (DSN) for ZS-SBIR. Specifically, a cross-modal contrastive method is proposed to learn generalized representations to smooth the domain gap by mining relations with additional augmented samples. Furthermore, a category-specific memory bank with sketch features is explored to reduce intra-class diversity in the sketch domain. Extensive experiments demonstrate that our approach notably outperforms the state-of-the-art methods in both Sketchy and TU-Berlin datasets.