Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Interactive Dynamic Video

interactivedynamicvideo.com

Post reply on HN
Print view View original

51–59 of 59 posts

Re: Interactive Dynamic Video

#51

Post by MichaelGG »

radarsat1 wrote: post #45
radarsat1 wrote: post #13

Dedicated host name for a single algorithm, TED talk, high-production value video with professional animations, narrator talking excitedly, more kitschy examples of applications than description of the algorithm itself. Yup, MIT.

Can't edit the comment so I'll just put this here: didn't expect this to get upvoted to be honest ;) I was trying to be bitter and petty as a kind of dark humour, but for the record I actually think it's pretty cool that students at MIT have that kind of support behind them. That said, it's always sort of amusing to see the MIT PR-machine at work, because it's so _obvious_ once you get familiar with how they work. At…

This is ridiculous. The amount of time that went into this project is probably fairly large. In a more competitive world, making some short videos and a basic page seems quite obvious. If the people involved spent, I dunno, 8 weeks on the project, then spent another few days putting together some materials to explain their work, that seems totally obvious and helpful.

Why do you think professionals were hired? A basic microphone+blanket to damp echos, and a passing familiarity (perhaps a friend) with a video editor should be enough to put that together. (No offense to the team.) If you wanted to hire someone, I'm betting you could get it done for $50 on Fiverr. Another $10 for a domain, and a few bucks for hosting.

You're making it sound like this is some huge professional outfit with a coordinated marketing plan. It looks more like someone trying to show off their project that they spent a lot of time on.

Re: Interactive Dynamic Video

#52

Post by abedavis »

erdevs wrote: post #50
abedavis wrote: post #46

Earlier quoted context omitted.

I am a grad student. If you consider how many hours I work, I probably make less than minimum wage. I am also the first author of the paper described in these videos, which my colleagues and I published in the top academic journal in our field. I also made all of the videos, did the voiceovers, and hand drew the animations (though I'm flattered you think they were professionally done). I wrote and presented the TED t…

I think your work (and your promotion of that work!) is awesome! Please take heart in all the appreciation and encouragement in this thread (and elsewhere)!

Thanks so much! I really appreciate that :) And, in fact, I'm not at all upset about the original comment - I totally get where the skepticism comes from. Polish and substance are too often uncorrelated when it comes to science in the media. I think it's important not to assume they are anti-correlated though, because then we start discouraging scientists from trying to reach people beyond the academic community.

Re: Interactive Dynamic Video

#53

Post by abedavis »

gavinpc wrote: post #10

What exactly are the implications of this for construction engineering, which must already have some way of measuring vibrations? I.e. does this measure something previously unmeasurable, or is it just that the visualization provides an extra channel for interpreting the data? Or is it that it would make this kind of tool available to anyone with a digital camera?

A big part of it is that measuring the vibrations of large structures, like buildings or bridges, can be inconvenient and expensive. There are limited cases where you can measure new stuff with video simply because you don't have to touch the object or shine a laser at it, but the big win is that by making vibration measurement cheap and simple/convenient, it means you can do a lot more of it. We can start taking more measurements on more structures, which will tell us where we should put resources to rebuild infrastructure.

It might also make quality assurance/ product testing cheaper but that's less save-the-world :-p

Re: Interactive Dynamic Video

#54

Post by icebraining »

mynewtb wrote: post #41
WhitneyLand wrote: post #26

Earlier quoted context omitted.

Jazzing up science seems to have more upside than not, as long as the quality and substance are maintained.

I recently got access to some hailed MIT technology with its own shiny videos, professional photos, patent pending and slick copy. It turned out to be a barely working prototype built on standard technology. To me misleading the public is not the worst about it. Pretending scientific accomplishments can deter and demotivate other, 'competing' scientists in the field.

Which tech was it?

Re: Interactive Dynamic Video

#55

Post by vanderZwan »

erdevs wrote: post #49
RIMR wrote: post #33

Earlier quoted context omitted.

Yes, my battery doesn't drain nearly fast enough when I'm playing. This would definitely fix that.

Funny comment. :) But actually this is unlikely to make a big difference on battery drain. Pokemon GO appears to be built on Unity and Unity is pretty notorious for having poor battery management. Whether you're pegging CPU or not, Unity burns lots of power. So, the incremental battery life hit is likely to be low in doing image processing on top of a frame.

Seriously, fuck Unity. It's the indy game equivalent of vendor-lock in, with everything negative about it.

They can't be arsed to fix a simple VSync bug on Linux, meaning even simple sprite games, or even more insulting: TIS-100, a game about optimising assembly on fake retro-hardware, deliberately made to look like a terminal, always burns 25% CPU power on my laptop.

More importantly, as Paolo Pedercini[0] mentioned on Twitter recently:

> I can open my Flash crap from 15 years ago no problem, but a project I made last year in Unity 5.1 crashes silently if rebuilt in Unity 5.3 I'm telling you: today's indie game production is a Unity monoculture that will disappear at the next hardware architecture cycle

[0] https://twitter.com/molleindustria/status/760904169444274176

Re: Interactive Dynamic Video

#56

Post by radarsat1 »

abedavis wrote: post #46
radarsat1 wrote: post #13

Dedicated host name for a single algorithm, TED talk, high-production value video with professional animations, narrator talking excitedly, more kitschy examples of applications than description of the algorithm itself. Yup, MIT.

I am a grad student. If you consider how many hours I work, I probably make less than minimum wage. I am also the first author of the paper described in these videos, which my colleagues and I published in the top academic journal in our field. I also made all of the videos, did the voiceovers, and hand drew the animations (though I'm flattered you think they were professionally done). I wrote and presented the TED t…

Sorry buddy! It was intended as a bit of dry humor. Didn't mean to pick a nerve. :) You're doing good work.

Re: Interactive Dynamic Video

#57

Post by abedavis »

spoiler wrote: post #36

This does look fascinating. The video is a bit showbusiness-like (for a lack of better expression). A few questions if the author is on HN: I wonder how long it takes to analyse a 5 second video? Also, it seems the algorithm only works with static images after the initial video analysis, or am I wrong? Also, how long does it take to render the "virtual" state of the object?

Hey, first author here. Good questions.

The simulation runs in realtime, and is interactive. Generating the simulation can take a while though.

The 5 second video took about a minute to process on my laptop. But longer videos, and higher-resolution videos, can take much longer. For instance, we've used the technique on high speed video to recover audible vibration modes, and this can take hours to process because the video itself is so big.

I wrote the processing code in Matlab, and the simulation in c++ and GLSL. The simulation is pretty well engineered because I wanted it to run in realtime. I didn't put a lot of time or effort into optimizing the video analysis part though, so it probably could be made faster.

Re: Interactive Dynamic Video

#58

Post by abedavis »

radarsat1 wrote: post #56
abedavis wrote: post #46

Earlier quoted context omitted.

I am a grad student. If you consider how many hours I work, I probably make less than minimum wage. I am also the first author of the paper described in these videos, which my colleagues and I published in the top academic journal in our field. I also made all of the videos, did the voiceovers, and hand drew the animations (though I'm flattered you think they were professionally done). I wrote and presented the TED t…

Sorry buddy! It was intended as a bit of dry humor. Didn't mean to pick a nerve. :) You're doing good work.

No worries. I totally understand (and invite) the skepticism. It's good to question this kind of press when you see it, and try to understand where the technology really is. I just think it's important not to assume that anything presented with polish has to be compromising on content.

Re: Interactive Dynamic Video

#59

Post by spoiler »

abedavis wrote: post #57
spoiler wrote: post #36

This does look fascinating. The video is a bit showbusiness-like (for a lack of better expression). A few questions if the author is on HN: I wonder how long it takes to analyse a 5 second video? Also, it seems the algorithm only works with static images after the initial video analysis, or am I wrong? Also, how long does it take to render the "virtual" state of the object?

Hey, first author here. Good questions. The simulation runs in realtime, and is interactive. Generating the simulation can take a while though. The 5 second video took about a minute to process on my laptop. But longer videos, and higher-resolution videos, can take much longer. For instance, we've used the technique on high speed video to recover audible vibration modes, and this can take hours to process because the…

Hi!

Thanks for responding to my questions! The whole project is amazing and I hope you keep working on this (and future projects) with the same enthusiasm as you do now! :D

Post reply on HN