Self-Training Boosted Multi-Factor Matching Network for Composed Image Retrieval

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wen, Haokun, Song, Xuemeng, Yin, Jianhua, Wu, Jianlong, Guan, Weili, Nie, Liqiang
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915037770153984
author Wen, Haokun
Song, Xuemeng
Yin, Jianhua
Wu, Jianlong
Guan, Weili
Nie, Liqiang
author_facet Wen, Haokun
Song, Xuemeng
Yin, Jianhua
Wu, Jianlong
Guan, Weili
Nie, Liqiang
contents The composed image retrieval (CIR) task aims to retrieve the desired target image for a given multimodal query, i.e., a reference image with its corresponding modification text. The key limitations encountered by existing efforts are two aspects: 1) ignoring the multi-faceted query-target matching factors; 2) ignoring the potential unlabeled reference-target image pairs in existing benchmark datasets. To address these two limitations is non-trivial due to the following challenges: 1) how to effectively model the multi-faceted matching factors in a latent way without direct supervision signals; 2) how to fully utilize the potential unlabeled reference-target image pairs to improve the generalization ability of the CIR model. To address these challenges, in this work, we first propose a muLtI-faceted Matching Network (LIMN), which consists of three key modules: multi-grained image/text encoder, latent factor-oriented feature aggregation, and query-target matching modeling. Thereafter, we design an iterative dual self-training paradigm to further enhance the performance of LIMN by fully utilizing the potential unlabeled reference-target image pairs in a semi-supervised manner. Specifically, we denote the iterative dual self-training paradigm enhanced LIMN as LIMN+. Extensive experiments on three real-world datasets, FashionIQ, Shoes, and Birds-to-Words, show that our proposed method significantly surpasses the state-of-the-art baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2305_09979
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Self-Training Boosted Multi-Factor Matching Network for Composed Image Retrieval
Wen, Haokun
Song, Xuemeng
Yin, Jianhua
Wu, Jianlong
Guan, Weili
Nie, Liqiang
Multimedia
The composed image retrieval (CIR) task aims to retrieve the desired target image for a given multimodal query, i.e., a reference image with its corresponding modification text. The key limitations encountered by existing efforts are two aspects: 1) ignoring the multi-faceted query-target matching factors; 2) ignoring the potential unlabeled reference-target image pairs in existing benchmark datasets. To address these two limitations is non-trivial due to the following challenges: 1) how to effectively model the multi-faceted matching factors in a latent way without direct supervision signals; 2) how to fully utilize the potential unlabeled reference-target image pairs to improve the generalization ability of the CIR model. To address these challenges, in this work, we first propose a muLtI-faceted Matching Network (LIMN), which consists of three key modules: multi-grained image/text encoder, latent factor-oriented feature aggregation, and query-target matching modeling. Thereafter, we design an iterative dual self-training paradigm to further enhance the performance of LIMN by fully utilizing the potential unlabeled reference-target image pairs in a semi-supervised manner. Specifically, we denote the iterative dual self-training paradigm enhanced LIMN as LIMN+. Extensive experiments on three real-world datasets, FashionIQ, Shoes, and Birds-to-Words, show that our proposed method significantly surpasses the state-of-the-art baselines.
title Self-Training Boosted Multi-Factor Matching Network for Composed Image Retrieval
topic Multimedia
url https://arxiv.org/abs/2305.09979