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
刚刚
Akim应助Ge采纳,获得10
刚刚
独特访天完成签到,获得积分10
1秒前
阿拉斯加今天学习了吗完成签到,获得积分10
2秒前
2秒前
科目三应助Zer采纳,获得10
2秒前
2秒前
3秒前
4秒前
嘟嘟噜完成签到,获得积分10
4秒前
5秒前
黑桃完成签到,获得积分10
5秒前
6秒前
6秒前
6秒前
7秒前
7秒前
7秒前
皮皮敏完成签到,获得积分20
7秒前
7秒前
7秒前
kun发布了新的文献求助10
8秒前
10秒前
10秒前
情怀应助江三村采纳,获得10
10秒前
洋洋发布了新的文献求助10
10秒前
10秒前
阿宝发布了新的文献求助10
10秒前
娇气的柜子完成签到,获得积分10
11秒前
vaco发布了新的文献求助10
11秒前
单薄的咖啡完成签到 ,获得积分10
12秒前
木语完成签到 ,获得积分10
12秒前
jhope应助天真蚂蚁采纳,获得10
12秒前
五档张诊人完成签到,获得积分10
14秒前
Tao发布了新的文献求助10
14秒前
111发布了新的文献求助10
14秒前
希望天下0贩的0应助kun采纳,获得10
15秒前
jeff发布了新的文献求助10
15秒前
研友_VZG7GZ应助娇气的柜子采纳,获得10
16秒前
16秒前
高分求助中
Spray / Wall-interaction Modelling by Dimensionless Data Analysis 2000
Write Like a Chemist: A Guide and Resource (第二版) 600
Mixed-anion Compounds 600
A new species of Velataspis (Hemiptera Coccoidea Diaspididae) from tea in Assam 500
Earth System Geophysics 500
Aspects of Babylonian celestial divination: the lunar eclipse tablets of Enūma Anu Enlil 500
Geochemistry, 2nd Edition 地球化学经典教科书第二版 401
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 3199835
求助须知:如何正确求助?哪些是违规求助? 2849046
关于积分的说明 8068740
捐赠科研通 2513318
什么是DOI,文献DOI怎么找? 1346068
科研通“疑难数据库(出版商)”最低求助积分说明 640167
邀请新用户注册赠送积分活动 609894