Results 21 to 30 of about 1,028,974 (354)
On the Integration of Self-Attention and Convolution [PDF]
Computer Vision and Pattern Recognition, 2021 Convolution and self-attention are two powerful techniques for representation learning, and they are usually considered as two peer approaches that are distinct from each other.
Xuran Pan +6 more
semanticscholar +1 more source
Algebra universalis, 2018 For a complete lattice $L$ and a relational structure $\mathfrak{X}=(X,(R_i)_I)$, we introduce the convolution algebra $L^{\mathfrak{X}}$. This algebra consists of the lattice $L^X$ equipped with an additional $n_i$-ary operation $f_i$ for each $n_i+1$-ary relation $R_i$ of $\mathfrak{X}$. For $ _1,\ldots, _{n_i}\in L^X$ and $x\in X$ we set $f_i( _1,
Elbert A. Walker +2 more
openaire +2 more sources
ECO: Efficient Convolution Operators for Tracking [PDF]
Computer Vision and Pattern Recognition, 2016 In recent years, Discriminative Correlation Filter (DCF) based methods have significantly advanced the state-of-the-art in tracking. However, in the pursuit of ever increasing tracking performance, their characteristic speed and real-time capability have
Martin Danelljan +3 more
semanticscholar +1 more source
Mathematical Biosciences and Engineering, 2023 Convolutional Neural Network (CNN) plays a vital role in the development of computer vision applications. The depth neural network composed of U-shaped structures and jump connections is widely used in various medical image tasks.
Xiaomeng Feng +5 more
doaj +1 more source
Free-Form Image Inpainting With Gated Convolution [PDF]
IEEE International Conference on Computer Vision, 2018 We present a generative image inpainting system to complete images with free-form mask and guidance. The system is based on gated convolutions learned from millions of images without additional labelling efforts. The proposed gated convolution solves the
Jiahui Yu +5 more
semanticscholar +1 more source
Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation [PDF]
Computer Vision and Pattern Recognition, 2020 State-of-the-art methods for large-scale driving-scene LiDAR segmentation often project the point clouds to 2D space and then process them via 2D convolution.
Xinge Zhu +7 more
semanticscholar +1 more source
DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement [PDF]
Interspeech, 2020 Speech enhancement has benefited from the success of deep learning in terms of intelligibility and perceptual quality. Conventional time-frequency (TF) domain methods focus on predicting TF-masks or speech spectrum, via a naive convolution neural network
Yanxin Hu +8 more
semanticscholar +1 more source
DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution [PDF]
Computer Vision and Pattern Recognition, 2020 Many modern object detectors demonstrate outstanding performances by using the mechanism of looking and thinking twice. In this paper, we explore this mechanism in the backbone design for object detection. At the macro level, we propose Recursive Feature
Siyuan Qiao, Liang-Chieh Chen, A. Yuille
semanticscholar +1 more source
Understanding Convolution for Semantic Segmentation [PDF]
IEEE Workshop/Winter Conference on Applications of Computer Vision, 2017 Recent advances in deep learning, especially deep convolutional neural networks (CNNs), have led to significant improvement over previous semantic segmentation systems.
Panqu Wang +6 more
semanticscholar +1 more source
Point cloud classification by dynamic graph CNN with adaptive feature fusion
IET Computer Vision, 2021 The deep neural network has made the most advanced breakthrough in almost all 2D image tasks, so we consider the application of deep learning in 3D images.
Rui Guo +6 more
doaj +1 more source