Self-Supervised Learning for RGB-Guided Depth Enhancement by Exploiting the Dependency Between RGB and Depth

RGB颜色模型 人工智能 计算机科学 依赖关系(UML) 计算机视觉 降噪 噪音(视频) 模式识别(心理学) 图像(数学)
作者
Jun Wang,Peilin Liu,Fei Wen
出处
期刊:IEEE transactions on image processing [Institute of Electrical and Electronics Engineers]
卷期号:32: 159-174 被引量:5
标识
DOI:10.1109/tip.2022.3226419
摘要

Due to the imaging mechanism of time-of-flight (ToF) sensors, the captured depth images usually suffer from severe noise and degradation. Though many RGB-guided methods have been proposed for depth image enhancement in the past few years, yet the enhancement performance on real-world depth images is still largely unsatisfactory. Two main reasons are the complexity of realistic noise and degradation in depth images, and the difficulty in collecting noise-clean pairs for supervised enhancement learning. This work aims to develop a self-supervised learning method for RGB-guided depth image enhancement, which does not require any noisy-clean pairs but can significantly boost the enhancement performance on real-world noisy depth images. To this end, we exploit the dependency between RGB and depth images to self-supervise the learning of the enhancement model. It is achieved by maximizing the cross-modal dependency between RGB and depth to promote the enhanced depth having dependency with the RGB of the same scene as much as possible. Furthermore, we augment the cross-modal dependency maximization formulation based on the optimal transport theory to achieve further performance improvement. Experimental results on both synthetic and real-world data demonstrate that our method can significantly outperform existing state-of-the-art methods on depth denoising, multi-path interference suppression, and hole filling. Particularly, our method shows remarkable superiority over existing ones on real-world data in handling various realistic complex degradation. Code is available at https://github.com/wjcyt/SRDE.
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
FashionBoy应助yan采纳,获得10
刚刚
易清华完成签到 ,获得积分10
1秒前
时光完成签到,获得积分10
3秒前
激昂的飞松完成签到,获得积分10
5秒前
轻松海云完成签到,获得积分10
8秒前
李爱国应助爱包的馒采纳,获得10
9秒前
小王同学完成签到 ,获得积分10
9秒前
愤怒的豌豆完成签到,获得积分10
14秒前
lianliyou完成签到,获得积分10
15秒前
三毛变相完成签到,获得积分10
17秒前
脑洞疼应助Cathay采纳,获得10
17秒前
阝火火完成签到 ,获得积分10
20秒前
JamesPei应助高兴的小采纳,获得10
22秒前
fawr完成签到 ,获得积分10
22秒前
乙酸乙酯会挥发完成签到,获得积分10
24秒前
雅2018完成签到 ,获得积分0
25秒前
phw2333应助Tonald Yang采纳,获得30
25秒前
fenghy完成签到,获得积分10
27秒前
如泣草芥完成签到,获得积分0
31秒前
王宇杰完成签到,获得积分10
31秒前
司马绮山完成签到,获得积分10
32秒前
我桽完成签到 ,获得积分10
33秒前
Maglev完成签到,获得积分10
36秒前
会撒娇的书白完成签到 ,获得积分10
38秒前
38秒前
摸鱼校尉完成签到,获得积分10
40秒前
专一的猎豹完成签到,获得积分10
40秒前
Ai香香完成签到,获得积分0
40秒前
shihui发布了新的文献求助30
41秒前
吗喽小祁完成签到,获得积分10
41秒前
在水一方应助皮皮鲁采纳,获得10
41秒前
书生也是小郎中完成签到 ,获得积分10
42秒前
爱科研的陈少完成签到,获得积分10
44秒前
小叶完成签到 ,获得积分10
44秒前
zlh完成签到 ,获得积分10
44秒前
汤汤发布了新的文献求助10
45秒前
还好发布了新的文献求助10
45秒前
49秒前
许诺完成签到,获得积分10
50秒前
54秒前
高分求助中
Spray / Wall-interaction Modelling by Dimensionless Data Analysis 2000
Mathematics and Finite Element Discretizations of Incompressible Navier—Stokes Flows 500
A new species of Coccus (Homoptera: Coccoidea) from Malawi 500
2-Acetyl-1-pyrroline: an important aroma component of cooked rice 500
A real-time energy management strategy based on fuzzy control and ECMS for PHEVs 400
Handbook on People's China (1957) 400
2024 Medicinal Chemistry Reviews 400
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 3190142
求助须知:如何正确求助?哪些是违规求助? 2839456
关于积分的说明 8023831
捐赠科研通 2502316
什么是DOI,文献DOI怎么找? 1336444
科研通“疑难数据库(出版商)”最低求助积分说明 637841
邀请新用户注册赠送积分活动 605999