Deep Multimodal Transfer Learning for Cross-Modal Retrieval

计算机科学 学习迁移 人工智能 模式 不相交集 模态(人机交互) 深度学习 知识转移 集合(抽象数据类型) 构造(python库) 领域(数学分析) 机器学习 情态动词 情报检索 自然语言处理 社会学 程序设计语言 高分子化学 化学 数学分析 组合数学 知识管理 社会科学 数学
作者
Liangli Zhen,Peng Hu,Xi Peng,Rick Siow Mong Goh,Joey Tianyi Zhou
出处
期刊:IEEE transactions on neural networks and learning systems [Institute of Electrical and Electronics Engineers]
卷期号:33 (2): 798-810 被引量:48
标识
DOI:10.1109/tnnls.2020.3029181
摘要

Cross-modal retrieval (CMR) enables flexible retrieval experience across different modalities (e.g., texts versus images), which maximally benefits us from the abundance of multimedia data. Existing deep CMR approaches commonly require a large amount of labeled data for training to achieve high performance. However, it is time-consuming and expensive to annotate the multimedia data manually. Thus, how to transfer valuable knowledge from existing annotated data to new data, especially from the known categories to new categories, becomes attractive for real-world applications. To achieve this end, we propose a deep multimodal transfer learning (DMTL) approach to transfer the knowledge from the previously labeled categories (source domain) to improve the retrieval performance on the unlabeled new categories (target domain). Specifically, we employ a joint learning paradigm to transfer knowledge by assigning a pseudolabel to each target sample. During training, the pseudolabel is iteratively updated and passed through our model in a self-supervised manner. At the same time, to reduce the domain discrepancy of different modalities, we construct multiple modality-specific neural networks to learn a shared semantic space for different modalities by enforcing the compactness of homoinstance samples and the scatters of heteroinstance samples. Our method is remarkably different from most of the existing transfer learning approaches. To be specific, previous works usually assume that the source domain and the target domain have the same label set. In contrast, our method considers a more challenging multimodal learning situation where the label sets of the two domains are different or even disjoint. Experimental studies on four widely used benchmarks validate the effectiveness of the proposed method in multimodal transfer learning and demonstrate its superior performance in CMR compared with 11 state-of-the-art methods.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
大幅提高文件上传限制,最高150M (2024-4-1)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
galeno完成签到 ,获得积分10
2秒前
隐形曼青应助zh采纳,获得30
3秒前
FashionBoy应助km采纳,获得30
3秒前
eisenchen发布了新的文献求助20
3秒前
3秒前
子铭发布了新的文献求助30
4秒前
FashionBoy应助闪电采纳,获得30
5秒前
过江春雷发布了新的文献求助10
7秒前
An完成签到,获得积分10
7秒前
泡泡发布了新的文献求助10
8秒前
华仔应助Qiyue采纳,获得30
10秒前
10秒前
慕青应助一只肥猫采纳,获得100
10秒前
11秒前
FashionBoy应助洛溪汐采纳,获得10
12秒前
无花果应助gyhmm采纳,获得10
13秒前
dddd完成签到,获得积分10
13秒前
13秒前
An发布了新的文献求助10
15秒前
15秒前
17秒前
Vroom发布了新的文献求助10
17秒前
无敌小行星完成签到,获得积分10
17秒前
17秒前
子云完成签到,获得积分10
17秒前
18秒前
liaoteng发布了新的文献求助10
18秒前
yiyiyi发布了新的文献求助10
18秒前
动人的水蓉完成签到,获得积分10
19秒前
19秒前
Zzz发布了新的文献求助10
21秒前
22秒前
老陈发布了新的文献求助10
25秒前
26秒前
26秒前
芷莯发布了新的文献求助10
26秒前
司空大有应助尼克采纳,获得10
26秒前
27秒前
lainka发布了新的文献求助10
27秒前
高分求助中
Sustainability in ’Tides Chemistry 2000
Sustainability in ’Tides Chemistry 1500
The ACS Guide to Scholarly Communication 1000
TM 5-855-1(Fundamentals of protective design for conventional weapons) 1000
Handbook of the Mammals of the World – Volume 3: Primates 805
Ethnicities: Media, Health, and Coping 800
胶体中的相变和自组装 500
热门求助领域 (近24小时)
化学 医学 生物 材料科学 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 基因 遗传学 催化作用 物理化学 免疫学 量子力学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 3071073
求助须知:如何正确求助?哪些是违规求助? 2725040
关于积分的说明 7488445
捐赠科研通 2372386
什么是DOI,文献DOI怎么找? 1257966
科研通“疑难数据库(出版商)”最低求助积分说明 610164
版权声明 596906