Vision Transformer for Contrastive Clustering

计算机科学 特征学习 聚类分析 人工智能 模式识别(心理学) 深度学习 卷积神经网络 自编码
作者
Hua-Bao Ling,Bin Zhu,Hui Dong,Ding-Hua Chen,Chang‐Dong Wang,Jianhuang Lai
出处
期刊:Cornell University - arXiv
标识
DOI:10.48550/arxiv.2206.12925
摘要

Vision Transformer (ViT) has shown its advantages over the convolutional neural network (CNN) with its ability to capture global long-range dependencies for visual representation learning. Besides ViT, contrastive learning is another popular research topic recently. While previous contrastive learning works are mostly based on CNNs, some recent studies have attempted to combine ViT and contrastive learning for enhanced self-supervised learning. Despite the considerable progress, these combinations of ViT and contrastive learning mostly focus on the instance-level contrastiveness, which often overlook the global contrastiveness and also lack the ability to directly learn the clustering result (e.g., for images). In view of this, this paper presents a novel deep clustering approach termed Vision Transformer for Contrastive Clustering (VTCC), which for the first time, to our knowledge, unifies the Transformer and the contrastive learning for the image clustering task. Specifically, with two random augmentations performed on each image, we utilize a ViT encoder with two weight-sharing views as the backbone. To remedy the potential instability of the ViT, we incorporate a convolutional stem to split each augmented sample into a sequence of patches, which uses multiple stacked small convolutions instead of a big convolution in the patch projection layer. By learning the feature representations for the sequences of patches via the backbone, an instance projector and a cluster projector are further utilized to perform the instance-level contrastive learning and the global clustering structure learning, respectively. Experiments on eight image datasets demonstrate the stability (during the training-from-scratch) and the superiority (in clustering performance) of our VTCC approach over the state-of-the-art.
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
隐形曼青应助zty123采纳,获得10
刚刚
1秒前
梅者如西发布了新的文献求助10
2秒前
顾矜应助十一采纳,获得10
3秒前
天天快乐应助封不迟采纳,获得10
4秒前
MaoTing完成签到,获得积分10
7秒前
9秒前
9秒前
10秒前
cyh413134发布了新的文献求助10
10秒前
11秒前
巧克力饼干完成签到 ,获得积分10
11秒前
11秒前
MaoTing发布了新的文献求助10
13秒前
慕青应助叶归采纳,获得10
13秒前
小王完成签到,获得积分10
14秒前
duosu发布了新的文献求助10
14秒前
李健应助羽羽采纳,获得10
15秒前
16秒前
lili发布了新的文献求助10
17秒前
zty123发布了新的文献求助10
17秒前
用头打碟完成签到,获得积分20
17秒前
踏实静枫完成签到,获得积分10
19秒前
finley123456完成签到 ,获得积分10
21秒前
窦逗豆完成签到,获得积分10
21秒前
小脑斧完成签到,获得积分10
21秒前
26秒前
26秒前
赘婿应助cyh413134采纳,获得10
28秒前
所所应助努力努力123采纳,获得10
28秒前
29秒前
31秒前
33秒前
34秒前
十一发布了新的文献求助10
36秒前
37秒前
鱼香rose盖饭完成签到,获得积分10
40秒前
文刀发布了新的文献求助10
41秒前
桐桐应助怡然平露采纳,获得10
42秒前
思源应助HouShipeng采纳,获得10
42秒前
高分求助中
Sustainable Land Management: Strategies to Cope with the Marginalisation of Agriculture 1000
Corrosion and Oxygen Control 600
Python Programming for Linguistics and Digital Humanities: Applications for Text-Focused Fields 500
Love and Friendship in the Western Tradition: From Plato to Postmodernity 500
Heterocyclic Stilbene and Bibenzyl Derivatives in Liverworts: Distribution, Structures, Total Synthesis and Biological Activity 500
重庆市新能源汽车产业大数据招商指南(两链两图两池两库两平台两清单两报告) 400
Division and square root. Digit-recurrence algorithms and implementations 400
热门求助领域 (近24小时)
化学 材料科学 医学 生物 有机化学 工程类 生物化学 纳米技术 物理 内科学 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 电极 光电子学 量子力学
热门帖子
关注 科研通微信公众号,转发送积分 2548718
求助须知:如何正确求助?哪些是违规求助? 2176691
关于积分的说明 5605713
捐赠科研通 1897439
什么是DOI,文献DOI怎么找? 946971
版权声明 565447
科研通“疑难数据库(出版商)”最低求助积分说明 503980