We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

This survey provides an overview of higher-order tensor decompositions, their applications, and available software. A tensor is a multidimensional or $N$-way array. Decompositions of higher-order tensors (i.e., $N$-way arrays with $N \geq 3$) have applications in psycho-metrics, chemometrics, signal processing, numerical linear algebra, computer vision, numerical analysis, data mining, neuroscience, graph analysis, and elsewhere. Two particular tensor decompositions can be considered to be higher-order extensions of the matrix singular value decomposition: CANDECOMP/PARAFAC (CP) decomposes a tensor as a sum of rank-one tensors, and the Tucker decomposition is a higher-order form of principal component analysis. There are many other tensor decompositions, including INDSCAL, PARAFAC2, CANDELINC, DEDICOM, and PARATUCK2 as well as nonnegative variants of all of the above. The N-way Toolbox, Tensor Toolbox, and Multilinear Engine are examples of software packages for working with tensors.

/pdf/tensor-decompositions-and-applications-46vl2ih5gn.pdf

Tensor Decompositions and Applications

The discriminative model learning for image denoising has been recently attracting considerable attentions due to its favorable denoising performance. In this paper, we take one step forward by investigating the construction of feed-forward denoising convolutional neural networks (DnCNNs) to embrace the progress in very deep architecture, learning algorithm, and regularization method into image denoising. Specifically, residual learning and batch normalization are utilized to speed up the training process as well as boost the denoising performance. Different from the existing discriminative denoising models which usually train a specific model for additive white Gaussian noise at a certain noise level, our DnCNN model is able to handle Gaussian denoising with unknown noise level (i.e., blind Gaussian denoising). With the residual learning strategy, DnCNN implicitly removes the latent clean image in the hidden layers. This property motivates us to train a single DnCNN model to tackle with several general image denoising tasks, such as Gaussian denoising, single image super-resolution, and JPEG image deblocking. Our extensive experiments demonstrate that our DnCNN model can not only exhibit high effectiveness in several general image denoising tasks, but also be efficiently implemented by benefiting from GPU computing.

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising

We present an approach to interpret the major surfaces, objects, and support relations of an indoor scene from an RGBD image. Most existing work ignores physical interactions or is applied only to tidy rooms and hallways. Our goal is to parse typical, often messy, indoor scenes into floor, walls, supporting surfaces, and object regions, and to recover support relationships. One of our main interests is to better understand how 3D cues can best inform a structured 3D interpretation. We also contribute a novel integer programming formulation to infer physical support relations. We offer a new dataset of 1449 RGBD images, capturing 464 diverse indoor scenes, with detailed annotations. Our experiments demonstrate our ability to infer support relations in complex scenes and verify that our 3D scene cues and inferred support lead to better object segmentation.

/pdf/indoor-segmentation-and-support-inference-from-rgbd-images-1brf1o9jc3.pdf

Indoor segmentation and support inference from RGBD images

Acquiring and representing the 4D space of rays in the world (the light field) is important for many computer vision and graphics applications. Yet, light field acquisition is costly due to their high dimensionality. Existing approaches either capture the 4D space explicitly, or involve an error-sensitive depth estimation process. This paper argues that the fundamental difference between different acquisition and rendering techniques is a difference between prior assumptions on the light field. We use the previously reported dimensionality gap in the 4D light field spectrum to propose a new light field prior. The new prior is a Gaussian assigning a non-zero variance mostly to a 3D subset of entries. Since there is only a low-dimensional subset of entries with non-zero variance, we can reduce the complexity of the acquisition process and render the 4D light field from 3D measurement sets. Moreover, the Gaussian nature of the prior leads to linear and depth invariant reconstruction algorithms. We use the new prior to render the 4D light field from a 3D focal stack sequence and to interpolate sparse directional samples and aliased spatial measurements. In all cases the algorithm reduces to a simple spatially invariant deconvolution which does not involve depth estimation.

/pdf/linear-view-synthesis-using-a-dimensionality-gap-light-field-57yjyh685b.pdf

Linear view synthesis using a dimensionality gap light field prior

When we take a picture through transparent glass the image we obtain is often a linear superposition of two images: the image of the scene beyond the glass plus the image of the scene reflected by the glass. Decomposing the single input image into two images is a massively ill-posed problem: in the absence of additional knowledge about the scene being viewed there are an infinite number of valid decompositions. In this paper we focus on an easier problem: user assisted separation in which the user interactively labels a small number of gradients as belonging to one of the layers.

/pdf/user-assisted-separation-of-reflections-from-a-single-image-qzxxbrlbbd.pdf

User Assisted Separation of Reflections from a Single Image Using a Sparsity Prior

Object motion during camera exposure often leads to noticeable blurring artifacts. Proper elimination of this blur is challenging because the blur kernel is unknown, varies over the image as a function of object velocity, and destroys high frequencies. In the case of motions along a 1D direction (e.g. horizontal) we show that these challenges can be addressed using a camera that moves during the exposure. Through the analysis of motion blur as space-time integration, we show that a parabolic integration (corresponding to constant sensor acceleration) leads to motion blur that is invariant to object velocity. Thus, a single deconvolution kernel can be used to remove blur and create sharp images of scenes with objects moving at different speeds, without requiring any segmentation and without knowledge of the object speeds. Apart from motion invariance, we prove that the derived parabolic motion preserves image frequency content nearly optimally. That is, while static objects are degraded relative to their image from a static camera, a reliable reconstruction of all moving objects within a given velocities range is made possible. We have built a prototype camera and present successful deblurring results over a wide variety of human motions.

/pdf/motion-invariant-photography-18l2z4nipy.pdf

Motion-invariant photography

Image restoration tasks are ill-posed problems, typically solved with priors. Since the optimal prior is the exact unknown density of natural images, actual priors are only approximate and typically restricted to small patches. This raises several questions: How much may we hope to improve current restoration results with future sophisticated algorithms? And more fundamentally, even with perfect knowledge of natural image statistics, what is the inherent ambiguity of the problem? In addition, since most current methods are limited to finite support patches or kernels, what is the relation between the patch complexity of natural images, patch size, and restoration errors? Focusing on image denoising, we make several contributions. First, in light of computational constraints, we study the relation between denoising gain and sample size requirements in a non parametric approach. We present a law of diminishing return, namely that with increasing patch size, rare patches not only require a much larger dataset, but also gain little from it. This result suggests novel adaptive variable-sized patch schemes for denoising. Second, we study absolute denoising limits, regardless of the algorithm used, and the converge rate to them as a function of patch size. Scale invariance of natural images plays a key role here and implies both a strictly positive lower bound on denoising and a power law convergence. Extrapolating this parametric law gives a ballpark estimate of the best achievable denoising, suggesting that some improvement, although modest, is still possible.

/pdf/patch-complexity-finite-pixel-correlations-and-optimal-39e6pcings.pdf

Patch complexity, finite pixel correlations and optimal denoising

When we take a picture through a window, the image we obtain is often a linear superposition of two images: the image of the scene beyond the window plus the image of the scene reflected by the window. Decomposing the single input image into two images is a massively ill-posed problem: in the absence of additional knowledge about the scene being viewed there is an infinite number of valid decompositions. We describe an algorithm that uses an extremely simple form of prior knowledge to perform the decomposition. Given a single image as input, the algorithm searches for a decomposition into two images that minimize the total amount of edges and comers. The search is performed using belief propagation on a patch representation of the image. We show that this simple prior is surprisingly powerful: our algorithm obtains "correct" separations on challenging reflection scenes using only a single image.

Anat Levin

Papers

Linear view synthesis using a dimensionality gap light field prior

User Assisted Separation of Reflections from a Single Image Using a Sparsity Prior

Motion-invariant photography

Patch complexity, finite pixel correlations and optimal denoising

Separating reflections from a single image using local features