#intermediate representation

[aneddoticamagazinestuff]

Everybody Dance Now: Motion Retargeting Video Subjects

New Post has been published on https://www.aneddoticamagazine.com/everybody-dance-now-motion-retargeting-video-subjects/

Everybody Dance Now: Motion Retargeting Video Subjects

“This paper presents a simple method for “do as I do” motion transfer: given a source video of a person dancing we can transfer that performance to a novel (amateur) target after only a few minutes of the target subject performing standard moves. We pose this problem as a per-frame image-to-image translation with spatio-temporal smoothing. Using pose detections as an intermediate representation between source and target, we learn a mapping from pose images to a target subject’s appearance. We adapt this setup for temporally coherent video generation including realistic face synthesis. Our video demo can be found here.” Caroline Chan, Shiry Ginosar, Tinghui Zhou, Alexei A. Efros  Website

[aneddoticamagazinestuff]

Everybody Dance Now: Motion Retargeting Video Subjects

New Post has been published on https://www.aneddoticamagazine.com/everybody-dance-now-motion-retargeting-video-subjects/

Everybody Dance Now: Motion Retargeting Video Subjects

“This paper presents a simple method for “do as I do” motion transfer: given a source video of a person dancing we can transfer that performance to a novel (amateur) target after only a few minutes of the target subject performing standard moves. We pose this problem as a per-frame image-to-image translation with spatio-temporal smoothing. Using pose detections as an intermediate representation between source and target, we learn a mapping from pose images to a target subject’s appearance. We adapt this setup for temporally coherent video generation including realistic face synthesis. Our video demo can be found here.” Caroline Chan, Shiry Ginosar, Tinghui Zhou, Alexei A. Efros  Website

  Related Post

Konrad Zuse and the first digital computer

Back to the Future Java

Children labour

“Rodriguez”: El año del gallo

IL FANTASMA DELL’ AUSTERITA’

Earth-Friendly Biofuel

A.(llucinazion)I. (parte seconda)

Swedish West Coast: Wave Energy Plant

Media Manipulation and Disinformation Online

WC in Cina

Xerox Alto Restoration playing Astroroids

Pete Kreiner: “Portrait of Ahed Tamimi / Fla...

.yuzo_related_post imgwidth:155px !important; height:145px !important; .yuzo_related_post .relatedthumbline-height:15px;background: !important;color:!important; .yuzo_related_post .relatedthumb:hoverbackground:#fcfcf4 !important; -webkit-transition: background 0.2s linear; -moz-transition: background 0.2s linear; -o-transition: background 0.2s linear; transition: background 0.2s linear;;color:!important; .yuzo_related_post .relatedthumb acolor:!important; .yuzo_related_post .relatedthumb a:hover color:!important;} .yuzo_related_post .relatedthumb:hover a color:!important; .yuzo_related_post .yuzo_text color:!important; .yuzo_related_post .relatedthumb:hover .yuzo_text color:!important; .yuzo_related_post .relatedthumb margin: 0px 0px 0px 0px; padding: 5px 5px 5px 5px; jQuery(document).ready(function( $ ) //jQuery('.yuzo_related_post').equalizer( overflow : 'relatedthumb' ); jQuery('.yuzo_related_post .yuzo_wraps').equalizer( columns : '> div' ); )

[aneddoticamagazinestuff]

Everybody Dance Now: Motion Retargeting Video Subjects

New Post has been published on https://www.aneddoticamagazine.com/everybody-dance-now-motion-retargeting-video-subjects/

Everybody Dance Now: Motion Retargeting Video Subjects

“This paper presents a simple method for “do as I do” motion transfer: given a source video of a person dancing we can transfer that performance to a novel (amateur) target after only a few minutes of the target subject performing standard moves. We pose this problem as a per-frame image-to-image translation with spatio-temporal smoothing. Using pose detections as an intermediate representation between source and target, we learn a mapping from pose images to a target subject’s appearance. We adapt this setup for temporally coherent video generation including realistic face synthesis. Our video demo can be found here.” Caroline Chan, Shiry Ginosar, Tinghui Zhou, Alexei A. Efros  Website

[z-bo]

Tuning applications for multi-core systems involve subtle concepts and target-dependent opti-mizations. New languages are being designed to express concurrency and locality without reference to a particular architecture. But compiling such abstractions into efficient code requires a portable, intermediate representation: this is essential for modular composition (separate compilation), for opti-mization frameworks independent of the source language, and for just-in-time compilation of bytecode languages. An intermediate representation is also essential to library and other baseline computing infrastructure developers who cannot afford the abstraction penalty of higher-level languages. But efficiency is nothing if it ruins productivity of the few available experts. Efficiency programmers need an alternative to fragile, ad-hoc optimizations built upon non-deterministic primitives. This paper introduces Erbium, an intermediate representation for compilers, a low-level language for efficiency programmers, and a lightweight runtime implementation. It is built upon a new data structure for scalable and deterministic concurrency, called Event Recording (Er). Our work is inspired by the semantics of data-flow and synchronous languages, motivated by advanced optimizations relying on non-blocking concurrency. We provide experimental evidence of the productivity, scalability and effi-ciency advantages of Erbium, relying on a prototype implementation in GCC4.3.

Wow!  I have been thinking about this exact topic for quite a while.