Related references
Note: Only part of the references are listed.
Article
Computer Science, Information Systems
Wenhui Zhou et al.
Summary: In this paper, a new camera ego-motion estimation method is proposed, focusing on the coupling of rotation and translation. A cascade decoupling structure is designed to separately learn the rotation and translation of camera motion. Meanwhile, a rigid-aware unsupervised learning framework is introduced to handle rigid motion and deformations in dynamic scenarios through joint learning of optical flow, stereo disparity, and camera pose.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Review
Optics
Bijie Bai et al.
Summary: Histological staining is an important technique in clinical pathology and research, but it is expensive, time-consuming, and limited in resource-limited settings. Deep learning techniques have provided a solution by digitally generating histological stains, which are rapid, cost-effective, and accurate alternatives to chemical staining methods. This review provides an overview of recent advances in deep learning-enabled virtual histological staining and discusses its future potential.
LIGHT-SCIENCE & APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Pradyumna Chari et al.
Summary: In this work, we teach a machine to detect the mechanical laws of motion using video and show its utility in computer vision tasks. The machine learns governing equations and parameters without prior knowledge of physics. We evaluate its performance using real and constructed videos, and demonstrate a real-world use case in object tracking where existing algorithms fail. Incorporating physics into computer vision not only serves curiosity-driven research but also provides an inductive bias for computer vision applications.
Article
Robotics
Alina Kloss et al.
Summary: This work explores the advantages and limitations of neural-network-based learning approaches for predicting the effects of physical interactions. It shows how analytical and learned models can be combined to leverage their respective strengths. A systematic evaluation on a large real-world dataset reveals that the hybrid architecture reduces required training data and improves generalization to novel physical interactions.
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH
(2022)
Article
Engineering, Industrial
Manuel Arias Chao et al.
Summary: A novel hybrid framework is proposed to combine physics-based performance models with deep learning algorithms for prognostics of complex safety-critical systems, improving prediction horizon by 127% compared to purely data-driven approaches. Physics-based performance models are used to infer unobservable model parameters related to system health and combined with sensor readings as input to a deep neural network, demonstrating superior performance over traditional data-driven methods.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Review
Behavioral Sciences
Celso M. de Melo et al.
Summary: Deep learning has achieved success in various domains, but the requirement for large amounts of labeled data presents a major bottleneck. Synthetic data is emerging as a potential solution, aided by advances in rendering pipelines, generative adversarial models, and fusion models. Domain adaptation techniques are also closing the statistical gap between synthetic and real data. The use of synthetic data and deep neural networks provides insights into the cognitive and neural functioning of biological systems.
TRENDS IN COGNITIVE SCIENCES
(2022)
Article
Computer Science, Software Engineering
A. Tewari et al.
Summary: Synthesizing photo-realistic images and videos is a key focus in computer graphics research. Neural rendering combines classical computer graphics techniques with machine learning to create algorithms for synthesizing images from real-world observations. This field has seen significant progress in recent years, with methods that can handle static scenes as well as non-rigidly deforming objects, scene editing, and composition. These methods have the advantage of being 3D-consistent and can be used for generative tasks. This report provides a comprehensive overview of state-of-the-art neural rendering methods, fundamental concepts, and open challenges.
COMPUTER GRAPHICS FORUM
(2022)
Article
Computer Science, Software Engineering
Alexander Vilesov et al.
Summary: With the rise of non-contact vital sign sensing during the COVID-19 pandemic, remote heart-rate monitoring has become increasingly important. However, previous studies have shown that using cameras can lead to a performance loss for individuals with darker skin tones. In this paper, the authors propose a solution by analyzing light transport and introducing a fairer modality - radar - for multi-modal fusion. The results show improved performance and fairness compared to existing methods, and a dataset with a focus on skin tone representation is made publicly available.
ACM TRANSACTIONS ON GRAPHICS
(2022)
Article
Computer Science, Software Engineering
G. Cai et al.
Summary: Mathematical representation of object shape is crucial for solving inverse rendering problems. Explicit representations are efficient for differentiable rendering but have difficulty handling topology changes. Implicit representations offer better support for topology changes but are harder to use for physics-based differentiable rendering. We introduce a new physics-based inverse rendering pipeline that utilizes both implicit and explicit representations. Our technique combines the benefits of both representations by supporting topology changes and differentiable rendering of complex effects.
COMPUTER GRAPHICS FORUM
(2022)
Article
Robotics
Isabella Huang et al.
Summary: Robotic grasping of 3D deformable objects is critical for various real-world applications. This study proposes studying the interaction with deformable objects through physics-based simulation and provides a simulated dataset and code repository for future research. The grasp outcomes on simulated objects show good correspondence with real counterparts.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Proceedings Paper
Engineering, Electrical & Electronic
Tzofi Klinghoffer et al.
Summary: This paper presents a framework to understand the building blocks of the emerging field of end-to-end design of camera hardware and algorithms, highlighting the transformation from physics-driven to data-driven and task-specific camera design. It emphasizes the prevalence of methods that combine both physics and data in imaging and computer vision.
2022 IEEE INTERNATIONAL CONFERENCE ON COMPUTATIONAL PHOTOGRAPHY (ICCP)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Chris Rockwell et al.
Summary: We propose a simple baseline for estimating the relative pose between two images, which can directly compute the rotation, translation, and scale. By making a few modifications to the Vision Transformer (ViT), we are able to achieve results close to the Eight-Point Algorithm. This approach provides a straightforward method that is highly competitive in various scenarios, especially in cases with limited data.
2022 INTERNATIONAL CONFERENCE ON 3D VISION, 3DV
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Yunhao Ba et al.
Summary: This research presents a large-scale dataset of real-world rainy and clean image pairs, and proposes a method to remove the degradations caused by rain streaks and accumulation. By collecting a real paired deraining dataset and using a robust deep neural network, the model outperforms existing deraining methods on real rainy images.
COMPUTER VISION, ECCV 2022, PT VII
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhen Wang et al.
Summary: Accelerated by telemedicine, advances in Remote Photoplethysmography (rPPG) are offering a feasible path for non-contact physiological measurement. However, limited datasets and lack of diversity in existing rPPG datasets result in accuracy disparities on different demographic groups. This paper proposes a biophysical learning method to generate physio-realistic synthetic rPPG videos and collects a diverse rPPG dataset to ensure healthcare equity.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Article
Computer Science, Interdisciplinary Applications
Boyuan Chen et al.
Summary: The article discusses how to determine the state variables of a system through high-dimensional observational data without prior knowledge. By proposing a principle and demonstrating its effectiveness in experiments, the article provides a new approach for identifying state variables.
NATURE COMPUTATIONAL SCIENCE
(2022)
Review
Computer Science, Artificial Intelligence
Bahram Jalali et al.
Summary: Physics has been successful in explaining nature using low-dimensional deterministic models, while artificial intelligence (AI) has achieved astonishing performance in domains like image classification and speech recognition through data-driven computational frameworks. However, AI's inconsistent predictions and computational complexity conflict with Moore's Law. This paper discusses how a symbiosis of physics and AI can overcome these challenges.
MACHINE LEARNING-SCIENCE AND TECHNOLOGY
(2022)
Article
Computer Science, Information Systems
Baigan Zhao et al.
Summary: This paper proposed a novel network for monocular VO problem that learns the latent subspace of optical flow and models sequential dynamics for motion estimation. By training the encoder separately in an unsupervised manner and using different network structures and training schemes, a more generalized and effective feature representation is achieved. Experiments on KITTI and Malaga datasets show that the LS-RCNN-VO model outperforms existing learning-based VO approaches.
Editorial Material
Multidisciplinary Sciences
Achuta Kadambi
Summary: Studying computer science can help ensure that medical devices are fair for all races and sexes.
Article
Computer Science, Artificial Intelligence
Jinshan Pan et al.
Summary: This study proposes an algorithm that addresses image restoration problems using generative models with adversarial learning, guided by physics models and trained in an end-to-end fashion for various low-level vision tasks, demonstrating superior performance compared to existing algorithms through extensive experiments.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Yeying Jin et al.
Summary: In this paper, an unsupervised network model DC-ShadowNet is proposed, which integrates a domain classifier to guide the generator and discriminator in handling shadow regions, while introducing novel loss functions. Experimental results demonstrate that the method is able to handle soft shadows and outperforms existing shadow removal methods in dealing with hard shadows.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Lina Liu et al.
Summary: A domain-separated network for self-supervised depth estimation of all-day images is proposed to alleviate the negative influence of disturbing terms, by partitioning into private and invariant domains, effectively tackling the illumination and domain shift between day and night images.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Automation & Control Systems
Arpit Agarwal et al.
Summary: This paper presents the first fully general optical tactile simulation system for a GelSight sensor using physics based rendering techniques. The proposed system outperforms previous simulation techniques in both qualitative and quantitative image similarity metrics. The code and experimental data are open-sourced on the project page.
2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021)
(2021)
Proceedings Paper
Automation & Control Systems
Yashraj Narang et al.
Summary: In this work, an efficient 3D finite element method (FEM) model of the SynTouch BioTac sensor was developed using an open-access, GPU-based robotics simulator, which achieved a speed 75 times faster than industry-standard, CPU-based simulator. Through self-supervision and latent representation learning, accurate synthesis of real-world BioTac electrical output and estimation of contact patches were achieved, even for unseen contact interactions.
2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Kai Zhang et al.
Summary: PhySG is an end-to-end inverse rendering pipeline that reconstructs geometry, materials, and illumination from images. It uses mixtures of spherical Gaussians and MLPs to represent specular BRDFs and geometry. The method is shown to work on scenes with challenging reflectance characteristics.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Zeyuan Chen et al.
Summary: Studies show that fine-tuning pre-trained models on synthetic data with real hazy images, combining multiple physical priors into a prior loss committee, significantly improves dehazing performance and achieves a new technological level in practical dehazing tasks.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Feilong Zhang et al.
Summary: This paper proposes a method that combines model-based alternative projection and deep neural networks for phase retrieval, aiming to achieve interpretability and effectiveness. The iterative process of phase retrieval is unfolded into a feed-forward neural network, embedding the physical model into its structure. Additionally, a complex-valued U-Net is proposed for image priori definition in dual planes.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Article
Environmental Sciences
Luca Schweri et al.
Summary: The accuracy of physical transport assessment is affected by noise in satellite-based wind retrievals and limited by sensor resolution. Reconstructing a continuous velocity field is crucial but challenging, with ambiguity due to missing visible clouds. The study demonstrates that a learning-based reconstruction method outperforms traditional models in handling large areas of missing data.
FRONTIERS IN CLIMATE
(2021)
Article
Computer Science, Artificial Intelligence
Wenhan Luo et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2020)
Correction
Optics
Christopher A. Metzler et al.
Article
Computer Science, Software Engineering
Seung-Hwan Baek et al.
ACM TRANSACTIONS ON GRAPHICS
(2020)
Article
Robotics
Andy Zeng et al.
IEEE TRANSACTIONS ON ROBOTICS
(2020)
Article
Computer Science, Software Engineering
Soshi Shimada et al.
ACM TRANSACTIONS ON GRAPHICS
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Thomas Nestmeyer et al.
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Simron Thapa et al.
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2020)
Article
Geochemistry & Geophysics
Janaki Vamaraju et al.
INTERPRETATION-A JOURNAL OF SUBSURFACE CHARACTERIZATION
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Shirsendu Sukanta Halder et al.
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Luyao Shi et al.
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2019, PT IV
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhuo Hui et al.
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Adam Rupe et al.
PROCEEDINGS OF 2019 5TH IEEE/ACM WORKSHOP ON MACHINE LEARNING IN HIGH PERFORMANCE COMPUTING ENVIRONMENTS (MLHPC 2019)
(2019)
Article
Computer Science, Artificial Intelligence
Matthias Muller et al.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
He Zhang et al.
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2018)
Review
Neurosciences
Demis Hassabis et al.
Article
Mathematical & Computational Biology
Adam H. Marblestone et al.
FRONTIERS IN COMPUTATIONAL NEUROSCIENCE
(2016)
Proceedings Paper
Computer Science, Artificial Intelligence
Yixin Zhu et al.
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2016)
Article
Multidisciplinary Sciences
Peter W. Battaglia et al.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2013)
Review
Computer Science, Artificial Intelligence
C. Goerzen et al.
JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS
(2010)
Review
Psychology, Developmental
Elizabeth S. Spelke et al.
DEVELOPMENTAL SCIENCE
(2007)
Article
Computer Science, Artificial Intelligence
Gary A. Atkinson et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2006)
Article
Engineering, Aerospace
E Frazzoli et al.
JOURNAL OF GUIDANCE CONTROL AND DYNAMICS
(2002)
Article
Computer Science, Artificial Intelligence
F Mokhtarian et al.
PATTERN RECOGNITION
(2002)
Article
Computer Science, Artificial Intelligence
LJ Latecki et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2000)
