Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Predicting depth is an essential component in understanding the 3D geometry of a scene. While for stereo images local correspondence suffices for estimation, finding depth relations from a single image is less straightforward, requiring integration of both global and local information from various cues. Moreover, the task is inherently ambiguous, with a large source of uncertainty coming from the overall scale. In this paper, we present a new method that addresses this task by employing two deep network stacks: one that makes a coarse global prediction based on the entire image, and another that refines this prediction locally. We also apply a scale-invariant error to help measure depth relations rather than scale. By leveraging the raw datasets as large sources of training data, our method achieves state-of-the-art results on both NYU Depth and KITTI, and matches detailed depth boundaries without the need for superpixelation.

/pdf/depth-map-prediction-from-a-single-image-using-a-multi-scale-260o56cfli.pdf

Depth Map Prediction from a Single Image using a Multi-Scale Deep Network

Let's read! We will often find out this sentence everywhere. When still being a kid, mom used to order us to always read, so did the teacher. Some books are fully read in a week and we need the obligation to support reading. What about now? Do you still love reading? Is reading only for you who have obligation? Absolutely not! We here offer you a new book enPDFd the perception of the visual world to read.

The Perception of the Visual World. By James J. Gibson. U.S.A.: Houghton Mifflin Company, 1950 (George Allen & Unwin, Ltd., London). Price 35s.

For many applications in graphics, design, and human computer interaction, it is essential to understand where humans look in a scene Where eye tracking devices are not a viable option, models of saliency can be used to predict fixation locations Most saliency approaches are based on bottom-up computation that does not consider top-down image semantics and often does not match actual eye movements To address this problem, we collected eye tracking data of 15 viewers on 1003 images and use this database as training and testing examples to learn a model of saliency based on low, middle and high-level image features This large database of eye tracking data is publicly available with this paper

/pdf/learning-to-predict-where-humans-look-20huj27145.pdf

Learning to predict where humans look

A morphable model for the synthesis of 3D faces

https://www.societyofrobots.com/robottheory/Biomimetic_robot_navigation.pdf

Biomimetic robot navigation

In recent years, kernel principal component analysis (KPCA) has been suggested for various image processing tasks requiring an image model such as, e.g., denoising or compression. The original form of KPCA, however, can be only applied to strongly restricted image classes due to the limited number of training examples that can be processed. We therefore propose a new iterative method for performing KPCA, the kernel Hebbian algorithm, which iteratively estimates the kernel principal components with only linear order memory complexity. In our experiments, we compute models for complex image classes such as faces and natural images which require a large number of training examples. The resulting image models are tested in single-frame super-resolution and denoising applications. The KPCA model is not specifically tailored to these tasks; in fact, the same model can be used in super-resolution with variable input resolution, or denoising with unknown noise characteristics, in spite of this, both super-resolution and denoising performance are comparable to existing methods.

/pdf/iterative-kernel-principal-component-analysis-for-image-1y8jesth52.pdf

Iterative kernel principal component analysis for image modeling

In homing tasks, the goal is often not marked by visible objects but must be inferred from the spatial relation to the visual cues in the surrounding scene. The exact computation of the goal direction would require knowledge about the distances to visible landmarks, information, which is not directly available to passive vision systems. However, if prior assumptions about typical distance distributions are used, a snapshot taken at the goal suffices to compute the goal direction from the current view. We show that most existing approaches to scene-based homing implicitly assume an isotropic landmark distribution. As an alternative, we propose a homing scheme that uses parameterized displacement fields. These are obtained from an approximation that incorporates prior knowledge about perspective distortions of the visual environment. A mathematical analysis proves that both approximations do not prevent the schemes from approaching the goal with arbitrary accuracy, but lead to different errors in the computed goal direction. Mobile robot experiments are used to test the theoretical predictions and to demonstrate the practical feasibility of the new approach.

Where did I take that snapshot? Scene-based homing by image matching

The human visual system is foveated, that is, outside the central visual ﬁeld resolution and acuity drop rapidly.Nonetheless much of a visual scene is perceived after only a few saccadic eye movements, suggesting an effectivestrategy for selecting saccade targets. It has been known for some time that local image structure at saccade targetsinﬂuences the selection process. However, the question of what the most relevant visual features are is still under debate.Here we show that center-surround patterns emerge as the optimal solution for predicting saccade targets from their localimage structure. The resulting model, a one-layer feed-forward network, is surprisingly simple compared to previouslysuggested models which assume much more complex computations such as multi-scale processing and multiple featurechannels. Nevertheless, our model is equally predictive. Furthermore, our ﬁndings are consistent with neurophysiologicalhardware in the superior colliculus. Bottom-up visual saliency may thus not be computed cortically as has been thoughtpreviously.Keywords: visual saliency, eye movements, receptive ﬁeld analysis, classiﬁcation images, kernel methods,support vector machines, natural scenesCitation: Kienzle, W., Franz, M. O., Scholkopf, B., & Wichmann, F. A. (2009). Center-surround patterns emerge as optimalpredictors for human saccade targets. Journal of Vision, 9(5):7, 1–15, http://journalofvision.org/9/5/7/, doi:10.1167/9.5.7.

Center-surround patterns emerge as optimal predictors for human saccade targets.

This paper addresses the bottom-up influence of local image information on human eye movements. Most existing computational models use a set of biologically plausible linear filters, e.g., Gabor or Difference-of-Gaussians filters as a front-end, the outputs of which are nonlinearly combined into a real number that indicates visual saliency. Unfortunately, this requires many design parameters such as the number, type, and size of the front-end filters, as well as the choice of nonlinearities, weighting and normalization schemes etc., for which biological plausibility cannot always be justified. As a result, these parameters have to be chosen in a more or less ad hoc way. Here, we propose to learn a visual saliency model directly from human eye movement data. The model is rather simplistic and essentially parameter-free, and therefore contrasts recent developments in the field that usually aim at higher prediction rates at the cost of additional parameters and increasing model complexity. Experimental results show that—despite the lack of any biological prior knowledge—our model performs comparably to existing approaches, and in fact learns image features that resemble findings from several previous studies. In particular, its maximally excitatory stimuli have center-surround structure, similar to receptive fields in the early human visual system.

/pdf/a-nonparametric-approach-to-bottom-up-visual-saliency-1mgil5soe2.pdf

Matthias O. Franz

Papers

Biomimetic robot navigation

Iterative kernel principal component analysis for image modeling

Where did I take that snapshot? Scene-based homing by image matching

Center-surround patterns emerge as optimal predictors for human saccade targets.

A Nonparametric Approach to Bottom-Up Visual Saliency