Interactive medical image segmentation methods have become increasingly popular in recent years. These methods combine manual lab.ling and automatic segmentation, reducing the workload of annotation while maintaining ...
详细信息
作者:
Guo, KuoLi, YifanChen, HaoShen, Hong-BinYang, YangShanghai Jiao Tong University
Key Lab. of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Department of Computer Science and Engineering Shanghai200240 China Shanghai Jiao Tong University
Key Laboratory of System Control and Information Processing Ministry of Education of China Institute of Image Processing and Pattern Recognition Shanghai200240 China Carnegie Mellon University
School of Computer Science Computational Biology Department PittsburghPA15213 United States
Isoforms refer to different mRNA molecules transcribed from the same gene, which can be translated into proteins with varying structures and functions. Predicting the functions of isoforms is an essential topic in bio...
详细信息
NOMA (Non-Orthogonal Multiple Access), as one of the candidate technologies of 5G, can improve the spectrum efficiency and system capacity, and has attracted wide attention. The essence of NOMA is multi-user overlay t...
详细信息
Urban rail transit (URT) is vulnerable to natural disasters and social emergencies including fire, storm and epidemic (such as COVID-19), and real-time origin-destination (OD) flow prediction provides URT operators wi...
详细信息
Bokeh effect transformation is a novel task in computer vision and computational photography. It aims to convert bokeh effects from one camera lens to another. To this end, we introduce a new concept of blur ratio, wh...
Bokeh effect transformation is a novel task in computer vision and computational photography. It aims to convert bokeh effects from one camera lens to another. To this end, we introduce a new concept of blur ratio, which represents the ratio of the blur amount of a target image to that of a source image, and propose a novel framework SBTNet based on this concept. For cat-eye simulation and lens type transformation, a two-channel coordinate map and a two-channel one-hot map are added as extra inputs. The core of the framework is a sequence of parallel FeaNets, along with a feature selection and integration strategy, which aims to transform the blur amount with arbitrary blur ratio. The effectiveness of the proposed framework is demonstrated through extensive experiments, and our solution has achieved the top LPIPS metric in NTIRE 2023 Bokeh Effect Transformation Challenge.
This paper develops a data-driven deterministic identification architecture for discovering stochastic differential equations (SDEs) directly from data. The architecture first generates deterministic data for stochast...
This paper develops a data-driven deterministic identification architecture for discovering stochastic differential equations (SDEs) directly from data. The architecture first generates deterministic data for stochastic processes using the Feynman–Kac formula, and gives a parabolic partial differential equation (PDE) associated with the SDE. Then, a sparse regression model is proposed to discover drift and diffusion terms in SDEs using PDE data-driven techniques, where a large candidate library of potential terms only for the drift and diffusion coefficients in SDEs need be constructed. To simultaneously infer the drift and diffusion terms, we proposed a sequential thresholded reweighted least-squares algorithm to solve the constructed sparse regression model. The main advantage of the proposed method is that on the one hand, theoretical and numerical identification results of PDEs can be used for SDEs, on the score, our SDE identification problem is translated into the parameter estimation problem of PDEs, on the other hand, the proposed algorithm is easily executed and can enhance the sparsity and accuracy. Through several classical SDEs and ordinary differential equations, the effectiveness of the proposed data-driven method is demonstrated, and several comparison experiments with state-of-the-art approaches is provided to illustrate the superiority of the developed algorithm.
Privacy-preserving image generation is particularly crucial in fields like healthcare, where data are both sensitive and limited. However, effective privacy preservation often compromises the visual quality and utilit...
详细信息
ISBN:
(数字)9798331516147
ISBN:
(纸本)9798331516154
Privacy-preserving image generation is particularly crucial in fields like healthcare, where data are both sensitive and limited. However, effective privacy preservation often compromises the visual quality and utility of the generated images due to privacy budget constraints. To address this issue, in this paper, We propose a novel network architecture, IRSEnet, which combines multi-scale feature extraction technology and residual channel attention mechanisms, aiming to enhance the visual quality of generated images and improve the performance of downstream classification tasks under differential privacy. The differential privacy mechanism ensures the security of sensitive data during training, while the multi-scale feature extraction module enhances feature extraction capabilities through parallel convolutional layers at multiple scales. Additionally, the channel attention module dynamically adjusts channel weights to focus on the most discriminative features. Experimental results demonstrate that this model significantly improves the utility of generated images and the accuracy of downstream classification tasks while preserving privacy. Future work will explore the application of this approach on larger datasets and across more diverse tasks.
We present 3D Cinemagraphy, a new technique that mar-ries 2D image animation with 3D photography. Given a single still image as input, our goal is to generate a video that contains both visual content animation and ca...
We present 3D Cinemagraphy, a new technique that mar-ries 2D image animation with 3D photography. Given a single still image as input, our goal is to generate a video that contains both visual content animation and camera motion. We empirically find that naively combining existing 2D image animation and 3D photography methods leads to obvious artifacts or inconsistent animation. Our key insight is that representing and animating the scene in 3D space offers a natural solution to this task. To this end, we first convert the input image into feature-based layered depth images using predicted depth values, followed by unprojecting them to a feature point cloud. To animate the scene, we perform motion estimation and lift the 2D motion into the 3D scene flow. Finally, to resolve the problem of hole emer-gence as points move forward, we propose to bidirectionally displace the point cloud as per the scene flow and synthe-size novel views by separately projecting them into target image planes and blending the results. Extensive experiments demonstrate the effectiveness of our method. A user study is also conducted to validate the compelling rendering results of our method.
In recent years, most of the studies have shown that the generalized iterated shrinkage thresholdings (GISTs) have become the commonly used first-order optimization algorithms in sparse learning problems. The nonconve...
详细信息
Recent research on human pose estimation exploits complex structures to improve performance on benchmark datasets, ignoring the resource overhead and inference speed when the model is actually deployed. In this paper,...
详细信息
暂无评论