itp: newportraits

[ITP: New Portraits] Final Project Update

in

Group members: Kay Wasil, Elif Ergin, and Cindy Hu

Idea

For our final project in Alan’s New Portraits class we need to make some volumetric captures using the Depthkit system, post-process, and make either a Unity project or some WEBGL assets for a live stream. This includes getting a subject, calibrating the systems, conducting interviews, lighting, and recording sound. Kay was super generous in opening up their community to us and they did a lot of hard work putting out an open call for drag performers to come into 370 and be our beautiful subjects!

Process

We booked out the Media Commons last Friday night to capture two performances. NYC-based performers Charlotte Harlotte and Sweaty Eddie generously shared their time and talents with us. We only had a 3 hour block of time so we had to move kind of quickly to make sure we could film everything we needed.

Calibration

First step is always calibrating the Depthkit system. We made super quick work of this the first time… but then we realized that the angles of the cameras were off because we couldn’t see feet through some of the camera views. Do the heavy cables pull the cameras down?

One question I do have is that after fixing the camera+sensor angles and then redoing the calibration in the original project, it seemed like the accuracy/precision went down. Does it break a project when adding new calibration points to an existing project? We decided to make a new project and start calibration over again, which seemed to work! Not sure if anyone recorded the final stats of our calibration but we got there eventually.

Lighting

After getting calibrated and checking all the camera views we moved onto lighting. There were still quite a few blue light flares which we worked on removing. This is also super weird because we fixed all the lighting in last class, right? Maybe the lights weren’t locked in place and gravity pulled them down too?

We set up a pink and orange light for Charlotte which looked SO good!

Our gorgeous performer Charlotte!

Sound

We wanted to record some interview questions so we set up a Zoom recorder. Here’s Elif while she’s really serious about the boom.

Filming

❤️❤️❤️

This experience was honestly so fun and super special! We were able to back up our files onto a hard drive and now I need to start reading up on the Unity setup.

[ITP: New Portraits] In Defense of the Poor Image

in

[Reading] In Defense of the Poor Image

Lol, what’s ironic is I had trouble downloading this image from the article…

A poor image is a preview, thumbnail, distributed for free, squeezed, compressed, reproduced, ripped, remixed, copied and pasted. As culture became a commodity and cinema was commercialized poor images actually made images/video/cinema/art accessible to the masses. Non-conformist and experimental visual matter disappeared from the surface into an underground of alternative archives and collections kept safe by humans. People who care enough to download and re-distribute images became co-authors. The medium of poor images is a co-created global network, a human network, distributed by word of mouth.

Here are some of my other notes:

  • “It transforms quality into accessibility, exhibition value into cult value, films into clips, contemplation into distraction.”

  • High-end/”rich” economies of film are rooted in national culture, capitalist studio production, male genius, conservative in their structure

  • lack of resolution = appropriation and displacement

  • poor image reveals the decline of experimental and non-commercial cinema

  • imperfect cinema = blurring the line between consumer and producer, audience and author

  • Users become editors, critics, translators, and (co-) authors of poor images —> popular images! For the people, people LOVE these images

Depthkit

We have our own Depthkit rig downstairs in the Media Commons. Depthkit it a software solution that captures volumetric video with a PC and depth sensors. The exported media is combined-per-pixel video files and are 3D objects at runtime in Unity.

Calibration

We have an appointment for the media commons 04/02 @ 4pm… 😈

[ITP: New Portraits] First Week Portraits!

in

3D scan portraits

For our first assignment, we had to make two 3D portraits, one of a person and one of an object, using the Polycam app. I don’t remember why that app didn’t work for me before, but now I use Scaniverse. Below are some screen shots of my portraits.

Kay

Okayyyy. So this portrait was kind of done in the spur of the moment. A “volumetric candid” if you will. But if I am being really honest, not much preparation of thought went into composing and posing this portrait. Kay and I sat next to each other on the ITP floor all Spring Break working through our stuff … and unfortunately this was kind of our vibe. Tired, burnt out, task-list oriented. Technically, tho, I think this turned out to be a pretty good scan!

Josh

Fail? Or portrait

Thinking of an object that might serve as a portrait, I was hoping to scan Josh’s phone and airpods. They are both “Josh yellow” and really personal items that allude to identity, personality, taste, etc.

When I tried scanning Josh’s phone and airpod case. My app refused to process the point cloud altogether, which has never happened to me before. My only guess is that it didn’t like the reflectivity of the phone screen? But that’s pretty weird because I’ve definitely scanned objects/spaces with mirrors before!

I tried again by scanning Josh’s water bottle this time. The scan came out way better than the previous one, except look at that weird bump! I think that this item serves as a much better portrait because I this object oozes Josh personality too.

[Reading] Sum of Profiles

physionotrace?!!?!

It is really crazy to think that artists found a way to make volumetric captures in the 1800’s! They developed a technology that works surprisingly similar to Polycam and Depthkit. They made sculptures using the photos of the different profiles of a subject and a master sculptor would carefully smooth the linear junctions between the carved profiles and unite them into a harmonious and just likeness of the subject. It was a “marriage of art and industry” and reminds me of the fact that portraiture wasn’t seen as a serious art form, only a craft, for a long time because of its mimetic nature.

Here are some technologies discussed in the paper:

  • Sequential construction - Rodin

  • Photo sculpture - Willeme

  • Mechanical sculpture/automatic sculpture - Willeme

  • Pantograph?

  • Physionotrace - Chrétien

  • “Gravure numismatique” - Collas

  • Smoke screen technology - Claudet

Advantages of this volumetric capture:

  • Microscopic subjects could be transformed into sculpture in very large proportions

  • Time saved and commensurate economic gain

  • “The advantage of the mechanical process was that it allowed the sculptor an amount of freedom to conceptualize and cultivate inspiration”

  • Realism

Resources

Sculpture as the Sum of Its Profiles: Fraincois Willeme and Photosculpture in France, 1859-1868