Définition d'écran | EPFL Graph Search

La définition d’écran est le nombre de points ou pixels (de l'anglais picture elements) que peut afficher un écran. La définition est le produit du nombre de points selon l’horizontale par le nombre de points selon la verticale de l’affichage. La définition est couramment confondue, à tort, avec la d'écran (la résolution prend en compte la taille de l'écran) ; un logiciel peut ainsi choisir parmi les différentes résolutions disponibles d'un écran. Pour utiliser une définition donnée, il faut que la carte graphique soit capable d'en générer le signal vidéo et que l'écran soit capable de l'afficher. La taille de l’écran et son pas de perçage (pitch) définissent la définition maximale que l’on peut atteindre sans perte de qualité en ce qui concerne l’image. Il est également possible d’utiliser plusieurs moniteurs avec une unité centrale. Cela permet d’additionner leurs nombres de pixels et ainsi d’utiliser par exemple une définition de . Pour un appareil photographique numérique, un

Novel small display using ferrofluid deformations as a visual information vector

Florent Caloz

Ferrofluids are suspensions of ferromagnetic nano-particles in a fluid carrier (typically oil). They have the property to change their shape when subjected to a magnetic field following the flux lines. Often used in design and art, the goal of this project is to develop a new kind of small screen (few cm) capable of displaying visual information using ferrofluid deformations. The aim is the design of a matrix of 15 pixels (3x5) capable of displaying distinct numbers and characters. Outer dimensions should not exceed 3x5 cm and the power consumption should be minimized. This project involves the design and optimization of specific electromagnets and the development of a control electronic board.

2011

AI-based telepresence for broadcast applications

Rayan Daod Nathoo, Touradj Ebrahimi

In this paper, we introduce a new solution and underlying architecture that allows remote participants to interact with hosts in a broadcast scenario. To achieve this, background extraction is first applied to video received from remote participants to extract their faces and bodies. Considering that the video from remote participants are usually of lower resolutions when compared to content produced by professional cameras in production, we propose to scale the extracted video with a super-resolution module. Finally, the processed video from remote participants are merged with studio video and streamed to audiences. Given the real-time and high-quality requirements, both background extraction and super-resolution modules are learning-based solutions and run on GPUs. The proposed solution has been deployed in the Advance Mixed Reality (AdMiRe) project. The objective and subjective assessment results show that the proposed solution works well in real world applications.

SPIE2022

Control of pulsed light propagation through multimode optical fibers

Edgar Emilio Morales Delgado

Visualization of organs and cells in the interior of living beings is a challenging task due to the light absorption and multiple light scattering occurring in biological tissue that prevents the di-rect transmission of images. A standard visualization approach is based on the use of endo-scopes which is accomplished through apertures in the body. Since their invention, the light transmission capability of optical fibers has enabled fiber optic based endoscopy with devices based on bundles of optical fibers that can directly relay images. Due to their small diameter, fiber optic endoscopes are standard in minimally invasive applications either with a white light illumination or as a fluorescent imaging device. The principle behind fluorescent endoscopy consists on the scanning of light intensity over a fluorescent sample, which is achieved by se-quential illumination of each one of the fiber bundles or by mechanical scanning of a single-mode double-cladding fiber. Collecting the fluorescent signal through the light guiding medium, an image can be reconstructed. The main drawback of fiber bundle endoscopes is the pixelated limited resolution given by the presence of thousands of cores that conform the bundle. In the double cladding fiber endoscopes, there is a limit to the miniaturization of the endoscope due to the requirement of mechanical elements at the tip of the fiber to scan the illumination over the sample. In this thesis, methods for focusing and digital scanning of optical light pulses through multimode optical fibers without any distal mechanical element are developed. Such techniques are based on spatial light modulation. Spatial phase modulation allows the control of light propagation in scrambling media, as in a multimode fiber, enabling the generation of intensity light foci or arbitrary intensity profiles. The high intensity of the transmitted pulses permits multi-photon imaging through multimode optical fibers, resulting in imaging probes of higher resolution than in the case of fiber bundles. We demonstrate in a working prototype that this approach provides all the advantages of multi-photon microscopy at the tip of an ultra-thin probe, such as a reduced photo bleaching of the sample, optical sectioning and enhanced image contrast. The developed methods can also be employed for material processing that requires high in-tensity light pulses. In specific, we demonstrate additive manufacturing, also known as 3D printing, at the tip of an ultra-thin needle. The working principle is based on two-photon polymerization, which is accomplished by scanning intense light pulses on a photosensitive material that hardens when exposed to light. All other additive manufacturing methods require very large nozzles or components in close proximity to the structure that they build, up to now. With an ultra-thin 3D printing probe we enable micro-fabrication through very small apertures, in places of difficult access or in the interior of living beings. We call it, endofabrication.

EPFL2017

Control of pulsed light propagation through multimode optical fibers

Edgar Emilio Morales Delgado

EPFL2017