This paper introduces a machine learning based system for controlling a robotic manipulator with visual perception only. The capability to autonomously learn robot controllers solely from raw-pixel images and without any prior knowledge of configuration is shown for the first time. We build upon the success of recent deep reinforcement learning and develop a system for learning target reaching ...

Sentence representation models trained only on language could potentially suffer from the grounding problem. Recent work has shown promising results in improving the qualities of sentence representations by jointly training them with associated image features. However, the grounding capability is limited due to distant connection between input sentences and image features by the design of the a...

We introduce a new sketch based interface for generating animations. Unlike traditional digital tools, ours is parameterized entirely by a neural network with no preprogrammed rules or knowledge representations. The capability of our sketching tool to support visual exploration and communication is demonstrated within the context of facial images, though our framework is domain independent. Our...

The digital recreation of real-world materials has a substantial role in applications such as product design, on-line shopping or video games. Since decisions in design or shopping are often driven by qualities like “softness” or “beautifulness” of a material (rather than its photo-accurate visual depiction), a digital material should not only closely capture the texture and reflectance of the ...

To evaluate changes in the visual processing of patients with progressive retinitis pigmentosa (RP) who acquired improved reading capability by eye-movement training (EMT), we performed functional magnetic resonance imaging (fMRI) before and after EMT. Six patients with bilateral concentric contraction caused by pigmentary degeneration of the retina and 6 normal volunteers were recruited. Patie...

A robot capable of understanding emotion expressions can increase its own capability of solving problems by using emotion expressions as part of its own decision-making, in a similar way to humans. Evidence shows that the perception of human interaction starts with an innate perception mechanism, where the interaction between different entities is perceived and categorized into two very clear d...

Here we show that bioluminescent organs of the squid Euprymna scolopes possess the molecular, biochemical, and physiological capability for light detection. Transcriptome analyses revealed expression of genes encoding key visual transduction proteins in light-organ tissues, including the same isoform of opsin that occurs in the retina. Electroretinograms demonstrated that the organ responds phy...

Attentive robots, inspired by human-like vision – are required to have visual systems with fovea-periphery distinction and saccadic motion capability. Thus, each frame in the incoming image sequence has nonuniform sampling and consecutive saccadic images have temporal redundancy. In this paper, we propose a novel video coding and streaming algorithm for low bandwidth networks that exploits thes...

A study of weight, auditory reaction time (ART), visual reaction time (VRT) and audiogram was carried out on 105 healthy female subjects between the age of 17-20 years during the premenstrual and post-menstrual phase. A significant increase in weight and prolongation in auditory reaction time (ART) and visual reaction time (VRT) were observed during premenstrual phase. A slight increase in deci...

the present study aims to investigate the system of the visual transitivity by analyzing the images of american english file (2014) and cutting edge (2005) series based on halliday’s (1976) systemic functional linguistics and kress and van leuween’s (1997-2006) social semiotics. the system of visual transitivity refers to a type of process which determines how represented participants are label...