Kernel (image processing)

About: Kernel (image processing) is a research topic. Over the lifetime, 12078 publications have been published within this topic receiving 238125 citations. The topic is also known as: convolution matrix & mask.

Aggregating Local Image Descriptors into Compact Codes

Abstract: This paper addresses the problem of large-scale image search. Three constraints have to be taken into account: search accuracy, efficiency, and memory usage. We first present and evaluate different ways of aggregating local image descriptors into a vector and show that the Fisher kernel achieves better performance than the reference bag-of-visual words approach for any given vector dimension. We then jointly optimize dimensionality reduction and indexing in order to obtain a precise vector comparison as well as a compact representation. The evaluation shows that the image representation can be reduced to a few dozen bytes while preserving high accuracy. Searching a 100 million image data set takes about 250 ms on one processor core.

Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring

Abstract: Non-uniform blind deblurring for general dynamic scenes is a challenging computer vision problem as blurs arise not only from multiple object motions but also from camera shake, scene depth variation. To remove these complicated motion blurs, conventional energy optimization based methods rely on simple assumptions such that blur kernel is partially uniform or locally linear. Moreover, recent machine learning based methods also depend on synthetic blur datasets generated under these assumptions. This makes conventional deblurring methods fail to remove blurs where blur kernel is difficult to approximate or parameterize (e.g. object motion boundaries). In this work, we propose a multi-scale convolutional neural network that restores sharp images in an end-to-end manner where blur is caused by various sources. Together, we present multi-scale loss function that mimics conventional coarse-to-fine approaches. Furthermore, we propose a new large-scale dataset that provides pairs of realistic blurry image and the corresponding ground truth sharp image that are obtained by a high-speed camera. With the proposed model trained on this dataset, we demonstrate empirically that our method achieves the state-of-the-art performance in dynamic scene deblurring not only qualitatively, but also quantitatively.

Background and foreground modeling using nonparametric kernel density estimation for visual surveillance

Abstract: Automatic understanding of events happening at a site is the ultimate goal for many visual surveillance systems. Higher level understanding of events requires that certain lower level computer vision tasks be performed. These may include detection of unusual motion, tracking targets, labeling body parts, and understanding the interactions between people. To achieve many of these tasks, it is necessary to build representations of the appearance of objects in the scene. This paper focuses on two issues related to this problem. First, we construct a statistical representation of the scene background that supports sensitive detection of moving objects in the scene, but is robust to clutter arising out of natural scene variations. Second, we build statistical representations of the foreground regions (moving objects) that support their tracking and support occlusion reasoning. The probability density functions (pdfs) associated with the background and foreground are likely to vary from image to image and will not in general have a known parametric form. We accordingly utilize general nonparametric kernel density estimation techniques for building these statistical representations of the background and the foreground. These techniques estimate the pdf directly from the data without any assumptions about the underlying distributions. Example results from applications are presented.

A Method for Optimal Image Subtraction

Abstract: We present a new method designed for optimal subtraction of two images with different seeing. Using image subtraction appears to be essential for full analysis of microlensing survey images; however, a perfect subtraction of two images is not easy, as it requires the derivation of an extremely accurate convolution kernel. Some empirical attempts to find the kernel have used a Fourier transform of bright stars, but solving the statistical problem of finding the best kernel solution has never really been tackled. We demonstrate that it is possible to derive an optimal kernel solution from a simple least-squares analysis using all the pixels of both images, and we also show that it is possible to fit the differential background variation at the same time. We show that point-spread function (PSF) variations can be easily handled by the method. To demonstrate the practical efficiency of the method, we analyzed some images from a Galactic Bulge field monitored by the OGLE II project. We find that the residuals in the subtracted images are very close to the photon noise expectations. We also present some light curves of variable stars and show that despite high crowding levels, we get an error distribution close to that expected from photon noise alone. We thus demonstrate that nearly optimal differential photometry can be achieved even in very crowded fields. We suggest that this algorithm might be particularly important for microlensing surveys, where the photometric accuracy and completeness levels could be very significantly improved by using this method.

Selective Kernel Networks

Abstract: In standard Convolutional Neural Networks (CNNs), the receptive fields of artificial neurons in each layer are designed to share the same size. It is well-known in the neuroscience community that the receptive field size of visual cortical neurons are modulated by the stimulus, which has been rarely considered in constructing CNNs. We propose a dynamic selection mechanism in CNNs that allows each neuron to adaptively adjust its receptive field size based on multiple scales of input information. A building block called Selective Kernel (SK) unit is designed, in which multiple branches with different kernel sizes are fused using softmax attention that is guided by the information in these branches. Different attentions on these branches yield different sizes of the effective receptive fields of neurons in the fusion layer. Multiple SK units are stacked to a deep network termed Selective Kernel Networks (SKNets). On the ImageNet and CIFAR benchmarks, we empirically show that SKNet outperforms the existing state-of-the-art architectures with lower model complexity. Detailed analyses show that the neurons in SKNet can capture target objects with different scales, which verifies the capability of neurons for adaptively adjusting their receptive field sizes according to the input. The code and models are available at https://github.com/implus/SKNet.