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

Computer vision has traditionally focused on extracting structure, such as depth, from images acquired using thin-lens or pinhole optics. The development of computational imaging is broadening this scope; a variety of unconventional cameras do not directly capture a traditional image anymore, but instead require the joint reconstruction of structure and image information. For example, recent coded aperture designs have been optimized to facilitate the joint reconstruction of depth and intensity. The breadth of imaging designs requires new tools to understand the tradeoffs implied by different strategies. This paper introduces a unified framework for analyzing computational imaging approaches. Each sensor element is modeled as an inner product over the 4D light field. The imaging task is then posed as Bayesian inference: given the observed noisy light field projections and a new prior on light field signals, estimate the original light field. Under common imaging conditions, we compare the performance of various camera designs using 2D light field simulations. This framework allows us to better understand the tradeoffs of each camera type and analyze their limitations.

/pdf/understanding-camera-trade-offs-through-a-bayesian-analysis-4e6hjvtmb1.pdf

Understanding camera trade-offs through a Bayesian analysis of light field projections - A revision

In this paper we study how estimates of ego-motion based on feature tracking (visual odometry) can be improved using a rough (low accuracy) map of where the observer has been. We call the process of aligning the visual ego-motion with the map locations as map correlation. Since absolute estimates of camera position are unreliable, we use stable local information such as change in orientation to perform the alignment. We also detect when the observer's path has crossed back on itself which helps improve both the visual odometry estimates and the alignment between the video and map sequences. The final alignment is computed using a graphical model whose MAP estimate is inferred using loopy belief propagation. Results are presented on a number of indoor and outdoor sequences.

/pdf/visual-odometry-and-map-correlation-2k2h8njjzp.pdf

Visual odometry and map correlation

Inferring internal scattering parameters for general, heterogeneous materials, remains a challenging inverse problem. Its difficulty arises from the complex way in which scattering materials interact with light, as well as the very high dimensionality of the material space implied by heterogeneity. The recent emergence of diverse computational imaging techniques, together with the widespread availability of computing power, present a renewed opportunity for tackling this problem. We take first steps in this direction, by deriving theoretical results, developing an algorithmic framework, and performing quantitative evaluations for the problem of heterogeneous inverse scattering from simulated measurements of different computational imaging configurations.

An Evaluation of Computational Imaging Techniques for Heterogeneous Inverse Scattering

We discuss the problem of ranking instances where an instance is associated with an integer from 1 to k. In other words, the specialization of the general multi-class learning problem when there exists an ordering among the instances — a problem known as “ordinal regression” or “ranking learning”. This problem arises in various settings both in visual recognition and other information retrieval tasks. In the context of applying a large margin principle to this learning problem, we introduce two main approaches for implementing the large margin optimization criteria for k margins. The first is the “fixed margin” policy in which the margin of the closest neighboring classes is being maximized — which turns out to be a direct generalization of SVM to ranking learning. The second approach allows for k different margins where the sum of margins is maximized, thus effectively having the solution biased towards the pairs of neighboring classes which are the farthest apart from each other. This approach is shown to reduce to SVM when the number of classes k . Both approaches are optimal in size (of the dual functional) of l where l is the total number of training examples. Experiments performed on visual classification and “collaborative filtering” show that both approaches outperform existing ordinal regression algorithms applied for ranking and multi-class SVM applied to general multi-class classification.

/pdf/taxonomy-of-large-margin-principle-algorithms-for-ordinal-hdmot2qjix.pdf

Taxonomy of Large Margin Principle Algorithms for Ordinal Regression Problems

Manifold pursuit extends principal component analysis to be invariant to a desired group of image-plane transformations of an ensemble of un-aligned images. We derive a simple technique for projecting a misaligned target image onto the linear subspace defined by the superpositions of a collection of model images. We show that it is possible to generate a fixed projection matrix which would separate the projected image into the aligned projected target and a residual image which accounts for the mis-alignment. An iterative procedure is then introduced for eliminating the residual image and leaving the correct aligned projected target image. Taken together, we demonstrate a simple and effective technique for obtaining invariance to image-plane transformations within a linear dimensionality reduction approach.

/pdf/manifold-pursuit-a-new-approach-to-appearance-based-j3ubuxxptl.pdf

Anat Levin

Papers

Understanding camera trade-offs through a Bayesian analysis of light field projections - A revision

Visual odometry and map correlation

An Evaluation of Computational Imaging Techniques for Heterogeneous Inverse Scattering

Taxonomy of Large Margin Principle Algorithms for Ordinal Regression Problems

Manifold pursuit: a new approach to appearance based recognition