Retrieval-Specific View Learning for Sketch-to-Shape Retrieval

计算机科学 观点 素描 人工智能 块(置换群论) 判别式 草图识别 自编码 联营 排名(信息检索) 任务(项目管理) 机器学习 模式识别(心理学) 深度学习 算法 艺术 手势识别 手势 几何学 数学 管理 经济 视觉艺术
作者
Shuaihang Yuan,Congcong Wen,Yu-Shen Liu,Yi Fang
出处
期刊:IEEE Transactions on Multimedia [Institute of Electrical and Electronics Engineers]
卷期号:: 1-12 被引量:3
标识
DOI:10.1109/tmm.2023.3287332
摘要

Sketch-based 3D shape retrieval (SBSR) can be approached by learning domain-invariant descriptors or ranking metrics from sketches and 2D view images of 3D shapes rendered through numerous viewpoints. However, determining the most appropriate viewpoints that convey discriminative geometric features to benefit the task of SBSR became an essential yet not fully explored area. Existing works extract 3D features from multi-view images observed through pre-defined viewpoints to match 2D sketches. Those methods, however, fail to dynamically select viewpoints by considering the SBSR task. In this work, we introduce a fully differentiable viewpoint learning paradigm driven by the downstream SBSR task, which supports the task-aware and sketch-dependent dynamic viewpoint determination process. We naturally integrate this task-specific and sketch-dependent viewpoint learning process into a meta-learning framework to develop a novel Dynamic Viewer (DV) module for SBSR. DV module comprises a Meta View Learner (MVL) block and a View Generator (VG) block. Specifically, as the first part of the DV module, the MVL block learns to initiate the necessary network parameters of the VG block. Then, the VG block that serves as the second part learns the best viewpoints to render 2D images. To learn the optimal viewpoints for SBSR, we further introduce a view mining loss that aims to maximize the similarity of feature-level information among rendered 2D views and the query sketch. Further, we adopt a variational autoencoder (VAE) to retrieve 3D shapes by setting the newly rendered images and query sketch as inputs. As evidenced by the comprehensive experimental results conducted on popular SBSR datasets, the proposed framework has been demonstrated to outperform recent methods in both category-level sketch-based and fine-grained SBSR.
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
香蕉觅云应助勤恳易谙采纳,获得10
1秒前
1秒前
fwx1997发布了新的文献求助10
2秒前
2秒前
2秒前
zf完成签到,获得积分10
3秒前
3秒前
cristin完成签到,获得积分10
3秒前
赵寇完成签到,获得积分10
4秒前
杨小勺儿发布了新的文献求助10
4秒前
玩命的朋友发布了新的文献求助100
4秒前
4秒前
燕儿发布了新的文献求助10
4秒前
4秒前
Novak关注了科研通微信公众号
5秒前
啥子那完成签到,获得积分10
5秒前
5秒前
李健应助柠檬味电子对儿采纳,获得10
6秒前
wayhome发布了新的文献求助10
6秒前
xiaxiao完成签到,获得积分0
6秒前
chen发布了新的文献求助10
6秒前
Elaine发布了新的文献求助10
6秒前
7秒前
8秒前
8秒前
luoyulin完成签到,获得积分10
9秒前
asstman发布了新的文献求助10
9秒前
orixero应助科研通管家采纳,获得10
9秒前
情怀应助科研通管家采纳,获得10
9秒前
英俊的铭应助科研通管家采纳,获得10
9秒前
隐形曼青应助科研通管家采纳,获得10
9秒前
HCLonely应助科研通管家采纳,获得10
9秒前
9秒前
9秒前
9秒前
9秒前
Xx完成签到,获得积分10
10秒前
10秒前
所所应助nlxx采纳,获得10
10秒前
无限的高烽完成签到,获得积分10
10秒前
高分求助中
Smart but Scattered: The Revolutionary Executive Skills Approach to Helping Kids Reach Their Potential (第二版) 1000
PraxisRatgeber: Mantiden: Faszinierende Lauerjäger 700
The Heath Anthology of American Literature: Early Nineteenth Century 1800 - 1865 Vol. B 500
A new species of Velataspis (Hemiptera Coccoidea Diaspididae) from tea in Assam 500
Machine Learning for Polymer Informatics 500
《关于整治突出dupin问题的实施意见》(厅字〔2019〕52号) 500
2024 Medicinal Chemistry Reviews 480
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 3221573
求助须知:如何正确求助?哪些是违规求助? 2870316
关于积分的说明 8170125
捐赠科研通 2537179
什么是DOI,文献DOI怎么找? 1369351
科研通“疑难数据库(出版商)”最低求助积分说明 645466
邀请新用户注册赠送积分活动 619101