DilateFormer: Multi-Scale Dilated Transformer for Visual Recognition

计算机科学 变压器 人工智能 冗余(工程) 利用 理论计算机科学 模式识别(心理学) 计算机视觉 物理 量子力学 操作系统 电压 计算机安全
作者
Jiayu Jiao,Yu-Ming Tang,Kun-Yu Lin,Yipeng Gao,J. Andy,Yaowei Wang,Wei‐Shi Zheng
出处
期刊:IEEE Transactions on Multimedia [Institute of Electrical and Electronics Engineers]
卷期号:25: 8906-8919 被引量:124
标识
DOI:10.1109/tmm.2023.3243616
摘要

As a de facto solution, the vanilla Vision Transformers (ViTs) are encouraged to model long-range dependencies between arbitrary image patches while the global attended receptive field leads to quadratic computational cost. Another branch of Vision Transformers exploits local attention inspired by CNNs, which only models the interactions between patches in small neighborhoods. Although such a solution reduces the computational cost, it naturally suffers from small attended receptive fields, which may limit the performance. In this work, we explore effective Vision Transformers to pursue a preferable trade-off between the computational complexity and size of the attended receptive field. By analyzing the patch interaction of global attention in ViTs, we observe two key properties in the shallow layers, namely locality and sparsity, indicating the redundancy of global dependency modeling in shallow layers of ViTs. Accordingly, we propose Multi-Scale Dilated Attention (MSDA) to model local and sparse patch interaction within the sliding window. With a pyramid architecture, we construct a Multi-Scale Dilated Transformer (DilateFormer) by stacking MSDA blocks at low-level stages and global multi-head self-attention blocks at high-level stages. Our experiment results show that our DilateFormer achieves state-of-the-art performance on various vision tasks. On ImageNet-1 K classification task, DilateFormer achieves comparable performance with 70% fewer FLOPs compared with existing state-of-the-art models. Our DilateFormer-Base achieves 85.6% top-1 accuracy on ImageNet-1 K classification task, 53.5% box mAP/46.1% mask mAP on COCO object detection/instance segmentation task and 51.1% MS mIoU on ADE20 K semantic segmentation task.
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
PDF的下载单位、IP信息已删除 (2025-6-4)

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
上官若男应助hugdoggy采纳,获得10
1秒前
emmm完成签到,获得积分10
1秒前
斯文败类应助anna采纳,获得10
2秒前
3秒前
默默的惜灵完成签到 ,获得积分10
3秒前
22发布了新的文献求助10
3秒前
Lucas应助pomfret采纳,获得10
5秒前
6秒前
6秒前
科目三应助娇气的天亦采纳,获得10
7秒前
JiangHan发布了新的文献求助10
7秒前
9秒前
GGBOND发布了新的文献求助10
9秒前
10秒前
黄振全发布了新的文献求助10
10秒前
11秒前
Lu发布了新的文献求助10
12秒前
苯氮小羊发布了新的文献求助10
12秒前
hugdoggy发布了新的文献求助10
14秒前
14秒前
英姑应助shawn采纳,获得10
17秒前
完美世界应助22采纳,获得10
17秒前
Hululu完成签到,获得积分10
17秒前
18秒前
精美礼物发布了新的文献求助30
19秒前
anna发布了新的文献求助10
19秒前
GGBOND发布了新的文献求助10
24秒前
pomfret完成签到,获得积分10
24秒前
从容甜瓜完成签到 ,获得积分10
26秒前
量子星尘发布了新的文献求助10
26秒前
27秒前
28秒前
桐桐应助Lu采纳,获得10
29秒前
YWang发布了新的文献求助10
31秒前
han应助归华采纳,获得10
32秒前
32秒前
SYLH应助zly采纳,获得30
35秒前
完美世界应助娇气的天亦采纳,获得10
37秒前
39秒前
42秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
‘Unruly’ Children: Historical Fieldnotes and Learning Morality in a Taiwan Village (New Departures in Anthropology) 400
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
基于可调谐半导体激光吸收光谱技术泄漏气体检测系统的研究 350
Robot-supported joining of reinforcement textiles with one-sided sewing heads 320
热门求助领域 (近24小时)
化学 材料科学 医学 生物 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 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3989115
求助须知:如何正确求助?哪些是违规求助? 3531367
关于积分的说明 11253688
捐赠科研通 3269986
什么是DOI,文献DOI怎么找? 1804868
邀请新用户注册赠送积分活动 882078
科研通“疑难数据库(出版商)”最低求助积分说明 809105