scholarly journals Sharpening of hierarchical visual feature representations of blurred images

2017 ◽  
Author(s):  
Mohamed Abdelhack ◽  
Yukiyasu Kamitani
Keyword(s):  
Neural Network ◽  
Visual System ◽  
Human Brain ◽  
Deep Neural Network ◽  
Prior Information ◽  
Visual Input ◽  
Top Down ◽  
Bottom Up ◽  
Reference Space ◽  
Blurred Images

AbstractThe robustness of the visual system lies in its ability to perceive degraded images. This is achieved through interacting bottom-up, recurrent, and top-down pathways that process the visual input in concordance with stored prior information. The interaction mechanism by which they integrate visual input and prior information is still enigmatic. We present a new approach using deep neural network (DNN) representation to reveal the effects of such integration on degraded visual inputs. We transformed measured human brain activity resulting from viewing blurred images to the hierarchical representation space derived from a feedforward DNN. Transformed representations were found to veer towards the original non-blurred image and away from the blurred stimulus image. This indicated deblurring or sharpening in the neural representation, and possibly in our perception. We anticipate these results will help unravel the interplay mechanism between bottom-up, recurrent, and top-down pathways, leading to more comprehensive models of vision.Significance statementOne powerful characteristic of the visual system is its ability to complement visual information for incomplete visual images. It operates by projecting information from higher visual and semantic areas of the brain into the lower and mid-level representations of the visual stimulus. We investigate the mechanism by which the human brain represents blurred visual stimuli. By decoding fMRI activity into a feedforward-only deep neural network reference space, we found that neural representations of blurred images are biased towards their corresponding deblurred images. This indicates a sharpening mechanism occurring in the visual cortex.

scholarly journals Deep Residual Network with Sparse Feedback for Image Restoration

2018 ◽  
Vol 8 (12) ◽  
pp. 2417 ◽  
Author(s):  
Zhenyu Guo ◽  
Yujuan Sun ◽  
Muwei Jian ◽  
Xiaofeng Zhang
Keyword(s):  
Neural Network ◽  
Image Restoration ◽  
Deep Neural Network ◽  
Convergence Speed ◽  
Fast Convergence ◽  
Experimental Results ◽  
Residual Network ◽  
Unknown Parameters ◽  
Noisy Images ◽  
Blurred Images

A deep neural network is difficult to train due to a large number of unknown parameters. To increase trainable performance, we present a moderate depth residual network for the restoration of motion blurring and noisy images. The proposed network has only 10 layers, and the sparse feedbacks are added in the middle and the last layers, which are called FbResNet. FbResNet has fast convergence speed and effective denoising performance. In addition, it can also reduce the artificial Mosaic trace at the seam of patches, and visually pleasant output results can be produced from the blurred images or noisy images. Experimental results show the effectiveness of our designed model and method.

scholarly journals Number detectors spontaneously emerge in a deep neural network designed for visual object recognition

2019 ◽  
Vol 5 (5) ◽  
pp. eaav7903 ◽  
Author(s):  
Khaled Nasr ◽  
Pooja Viswanathan ◽  
Andreas Nieder
Keyword(s):  
Neural Network ◽  
Object Recognition ◽  
Visual System ◽  
Deep Neural Network ◽  
Number Sense ◽  
Discrimination Performance ◽  
Visual Object ◽  
Visual Object Recognition ◽  
Biologically Inspired ◽  
Number Discrimination

Humans and animals have a “number sense,” an innate capability to intuitively assess the number of visual items in a set, its numerosity. This capability implies that mechanisms to extract numerosity indwell the brain’s visual system, which is primarily concerned with visual object recognition. Here, we show that network units tuned to abstract numerosity, and therefore reminiscent of real number neurons, spontaneously emerge in a biologically inspired deep neural network that was merely trained on visual object recognition. These numerosity-tuned units underlay the network’s number discrimination performance that showed all the characteristics of human and animal number discriminations as predicted by the Weber-Fechner law. These findings explain the spontaneous emergence of the number sense based on mechanisms inherent to the visual system.

scholarly journals Characterizing the effects of feature salience and top-down attention in the early visual system

2017 ◽  
Vol 118 (1) ◽  
pp. 564-573 ◽  
Author(s):  
Sonia Poltoratski ◽  
Sam Ling ◽  
Devin McCormack ◽  
Frank Tong
Keyword(s):  
Visual System ◽  
Spatial Attention ◽  
Visual Processing ◽  
Visual Pathway ◽  
Additive Effects ◽  
Voluntary Attention ◽  
Top Down ◽  
Motion Direction ◽  
Bottom Up ◽  
Visual Areas

The visual system employs a sophisticated balance of attentional mechanisms: salient stimuli are prioritized for visual processing, yet observers can also ignore such stimuli when their goals require directing attention elsewhere. A powerful determinant of visual salience is local feature contrast: if a local region differs from its immediate surround along one or more feature dimensions, it will appear more salient. We used high-resolution functional MRI (fMRI) at 7T to characterize the modulatory effects of bottom-up salience and top-down voluntary attention within multiple sites along the early visual pathway, including visual areas V1–V4 and the lateral geniculate nucleus (LGN). Observers viewed arrays of spatially distributed gratings, where one of the gratings immediately to the left or right of fixation differed from all other items in orientation or motion direction, making it salient. To investigate the effects of directed attention, observers were cued to attend to the grating to the left or right of fixation, which was either salient or nonsalient. Results revealed reliable additive effects of top-down attention and stimulus-driven salience throughout visual areas V1–hV4. In comparison, the LGN exhibited significant attentional enhancement but was not reliably modulated by orientation- or motion-defined salience. Our findings indicate that top-down effects of spatial attention can influence visual processing at the earliest possible site along the visual pathway, including the LGN, whereas the processing of orientation- and motion-driven salience primarily involves feature-selective interactions that take place in early cortical visual areas. NEW & NOTEWORTHY While spatial attention allows for specific, goal-driven enhancement of stimuli, salient items outside of the current focus of attention must also be prioritized. We used 7T fMRI to compare salience and spatial attentional enhancement along the early visual hierarchy. We report additive effects of attention and bottom-up salience in early visual areas, suggesting that salience enhancement is not contingent on the observer’s attentional state.

scholarly journals Interaction between Bottom-up Saliency and Top-down Control: How Saliency Maps Are Created in the Human Brain

2012 ◽  
Vol 22 (12) ◽  
pp. 2943-2952 ◽  
Author(s):  
Lucia Melloni ◽  
Sara van Leeuwen ◽  
Arjen Alink ◽  
Notger G. Müller
Keyword(s):  
Human Brain ◽  
Top Down ◽  
Bottom Up ◽  
Saliency Maps

scholarly journals Ramsey’s Theory Meets The Human Brain Connectome

2021 ◽  
Author(s):  
Arturo Tozzi
Keyword(s):  
Neural Networks ◽  
Human Brain ◽  
Network Theory ◽  
Neuronal Networks ◽  
Top Down ◽  
Bottom Up ◽  
Functional Interactions ◽  
Brain Connectome

Ramsey’s theory (RAM) from combinatorics and network theory goes looking for regularities and repeated patterns inside structures equipped with nodes and edges. RAM represents the outcome of a dual methodological commitment: by one side a top-down approach evaluates the possible arrangement of specific subgraphs when the number of graph’s vertices is already known, by another side a bottom-up approach calculates the possible number of graph’s vertices when the arrangement of specific subgraphs is already known. Since natural neural networks are often represented in terms of graphs, we suggest to utilize RAM for the analytical and computational assessment of a peculiar structure supplied with neuronal vertices and axonal edges, i.e., the human brain connectome. We discuss how a RAM approach in neuroscientific issues might be able to locate and trace unexplored motifs shared between different cortical and subcortical subareas. Furthermore, we will describe how notable RAM outcomes, such as the Ramsey’s theorem and the Ramsey’s number, could contribute to uncover still unknown anatomical connexions endowed in neuronal networks and unexpected functional interactions among grey zones of the human brain.

Sign in / Sign up

Export Citation Format

Share Document