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
fifteen发布了新的文献求助10
2秒前
轩仔发布了新的文献求助20
3秒前
单薄广山发布了新的文献求助10
4秒前
Leo发布了新的文献求助10
4秒前
5秒前
哎哟很烦完成签到,获得积分10
5秒前
6秒前
garbage完成签到 ,获得积分10
9秒前
今后应助Scarlett采纳,获得10
10秒前
WSGQT发布了新的文献求助10
11秒前
不配.应助Leo采纳,获得10
13秒前
科目三应助longdechuanren采纳,获得20
13秒前
Hello应助研友_ZAxX6n采纳,获得10
13秒前
深情安青应助66666天采纳,获得10
13秒前
conker完成签到,获得积分10
15秒前
fifteen应助菲菲采纳,获得10
15秒前
15秒前
酷波er应助nujabes采纳,获得10
16秒前
16秒前
yyymmma应助whims采纳,获得10
16秒前
yyymmma应助whims采纳,获得10
16秒前
yyymmma应助whims采纳,获得10
17秒前
yyymmma应助whims采纳,获得10
17秒前
yyymmma应助whims采纳,获得10
17秒前
孤檠应助whims采纳,获得10
17秒前
yyymmma应助whims采纳,获得10
17秒前
大模型应助whims采纳,获得10
17秒前
yyymmma应助whims采纳,获得10
17秒前
yyymmma应助whims采纳,获得10
17秒前
敏感的缘郡完成签到 ,获得积分10
17秒前
18秒前
20秒前
fifteen应助huangxihui采纳,获得10
21秒前
Suibobobo发布了新的文献求助10
21秒前
nowfitness完成签到,获得积分10
22秒前
田様应助whims采纳,获得10
22秒前
勤奋幻柏发布了新的文献求助10
22秒前
善学以致用应助whims采纳,获得10
22秒前
CodeCraft应助whims采纳,获得10
23秒前
赘婿应助whims采纳,获得10
23秒前
高分求助中
Spray / Wall-interaction Modelling by Dimensionless Data Analysis 2000
Evolution 3rd edition 1500
ALA生合成不全マウスでの糖代謝異常の分子機構解析 520
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
Handbook on People's China (1957) 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 3189013
求助须知:如何正确求助?哪些是违规求助? 2838460
关于积分的说明 8019847
捐赠科研通 2501314
什么是DOI,文献DOI怎么找? 1335507
科研通“疑难数据库(出版商)”最低求助积分说明 637555
邀请新用户注册赠送积分活动 605656