1,721,176 research outputs found
How much can orientation selectivity and contrast gain control reduce redundancies in natural images.
No full textThe two most prominent features of early visual processing are orientation selective filtering and
contrast gain control. While the effect of orientation selectivity can be assessed within in a linear model, contrast gain control is an inherently nonlinear computation. Here we employ the class of Lp elliptically contoured distributions to investigate the extent to which the two features, orientation selectivity and contrast gain control, are suited to model the statistics of natural images. Within this model we find that contrast gain control can play a significant role for the removal of redundancies in natural images. Orientation selectivity, in contrast, has only a very limited potential for linear redundancy reduction
One-Shot Segmentation in Clutter
No full textWe tackle the problem of one-shot segmentation: finding and segmenting a previously unseen object in a cluttered scene based on a single instruction example. We propose a novel dataset, which we call . Using a baseline architecture combining a Siamese embedding for detection with a U-net for segmentation we show that increasing levels of clutter make the task progressively harder. Using oracle models with access to various amounts of ground-truth information, we evaluate different aspects of the problem and show that in this kind of visual search task, detection and segmentation are two intertwined problems, the solution to each of which helps solving the other. We therefore introduce , an improved model that attends to multiple candidate locations, generates segmentation proposals to mask out background clutter and selects among the segmented objects. Our findings suggest that such image recognition models based on an iterative refinement of object detection and foreground segmentation may provide a way to deal with highly cluttered scenes
How Much Can Orientation Selectivity and Contrast Gain Control Reduce the Redundancies in Natural Images
Full text linkThe two most prominent features of early visual processing are orientation selective filtering and contrast gain control. While the effect of orientation selectivity can be assessed within in a linear model, contrast gain control is an inherently nonlinear computation. Here we employ the class of L_p elliptically contoured distributions to investigate the extent to which the two features, orientation selectivity and contrast gain control, are suited to model the statistics of natural images. Within this model we find that contrast gain control can play a significant role for the removal of redundancies in natural images. Orientation selectivity, in contrast, has only a very limited potential for linear redundancy reduction
What is the limit of redundancy reduction with divisive normalization?
No full textDivisive normalization has been proposed as a nonlinear redundancy reduction mechanism capturing contrast correlations. Its basic function is a radial rescaling of the population response. Because of the saturation of divisive normalization, however, it is impossible to achieve a fully independent representation. In this letter, we derive an analytical upper bound on the inevitable residual redundancy of any saturating radial rescaling mechanism
Neural system identification for large populations separating "what" and "where"
No full textNeuroscientists classify neurons into different types that perform similar computations at different locations in the visual field. Traditional methods for neural system identification do not capitalize on this separation of 'what' and 'where'. Learning deep convolutional feature spaces that are shared among many neurons provides an exciting path forward, but the architectural design needs to account for data limitations: While new experimental techniques enable recordings from thousands of neurons, experimental time is limited so that one can sample only a small fraction of each neuron's response space. Here, we show that a major bottleneck for fitting convolutional neural networks (CNNs) to neural data is the estimation of the individual receptive field locations, a problem that has been scratched only at the surface thus far. We propose a CNN architecture with a sparse readout layer factorizing the spatial (where) and feature (what) dimensions. Our network scales well to thousands of neurons and short recordings and can be trained end-to-end. We evaluate this architecture on ground-truth data to explore the challenges and limitations of CNN-based system identification. Moreover, we show that our network model outperforms current state-of-the art system identification models of mouse primary visual cortex
SALAD: A Toolbox for Semi-supervised Adaptive Learning Across Domains
No full textWe introduce salad, an open source toolbox that provides a unified implementation of state-of-the-art methods for transfer learning, semi-supervised learning and domain adaptation. In the first release, we provide a framework for reproducing, extending and combining research results of the past years, including model architectures, loss functions and training algorithms. The toolbox along with first benchmark results and further resources is accessible at domainadaptation.org
Multi-Task Generalization and Adaptation between Noisy Digit Datasets: An Empirical Study
No full textTransfer learning for adaptation to new tasks is usually performed by either finetuning all model parameters or parameters in the final layers. We show that good target performance can also be achieved on typical domain adaptation tasks by adapting only the normalization statistics and affine transformations of layers throughout the network. We apply this adaptation scheme to supervised domain adaptation on common digit datasets and study robustness properties under perturbation by noise. Our results indicate that (1) adaptation to noise exceeds the difficulty of widely used digit benchmarks in domain adaptation,(2) the similarity of the optimal adaptation parameters for different domains is strongly predictive of generalization performance, and (3) generalization performance is highest with training on a rich environment or high noise levels
Texture Modelling Using Convolutional Neural Networks
No full textWe introduce a new model of natural textures based on the feature spaces of convolutional neural networks optimised for object recognition. Samples from the model are of high perceptual quality demonstrating the generative power of neural networks trained in a purely discriminative fashion. Within the model, textures are represented by the correlations between feature maps in several layers of the network. We show that across layers the texture representations increasingly capture the statistical properties of natural images while making object information more and more explicit. Extending this framework to texture transfer, we introduce A Neural Algorithm of Artistic Style that can separate and recombine the image content and style of natural images. The algorithm allows us to produce new artistic imagery that combines the content of an arbitrary photograph with the appearance of numerous well-known artworks, thus offering a path towards an algorithmic understanding of how humans create and perceive artistic imagery
One-Shot Instance Segmentation
No full textWe tackle the problem of one-shot instance segmentation: Given an example image of a novel, previously unknown object category, find and segment all objects of this category within a complex scene. To address this challenging new task, we propose Siamese Mask R-CNN. It extends Mask R-CNN by a Siamese backbone encoding both reference image and scene, allowing it to target detection and segmentation towards the reference category. We demonstrate empirical results on MS Coco highlighting challenges of the one-shot setting: while transferring knowledge about instance segmentation to novel object categories works very well, targeting the detection network towards the reference category appears to be more difficult. Our work provides a first strong baseline for one-shot instance segmentation and will hopefully inspire further research into more powerful and flexible scene analysis algorithms. Code is available at: https://github.com/bethgelab/siamese-mask-rcn
SALAD: A Toolbox for Semi-supervised Adaptive Learning Across Domains
No full textWe introduce salad, an open source toolbox that provides a unified implementation of state-of-the-art methods for transfer learning, semi-supervised learning and domain adaptation. In the first release, we provide a framework for reproducing, extending and combining research results of the past years, including model architectures, loss functions and training algorithms. The toolbox along with first benchmark results and further resources is accessible at domainadaptation.org
- …
