-
Shared Minds Dimensionality (performance) Thesis Video in Dumbo New Portraits Synthetic Architectures Reconstructing Lost Spaces Web Art as Site
yilin ye
About
Portfolio
GitHub
yilin ye
About
Portfolio
GitHub
Performance @IDM Show at The Yard & @ITP Show
Show days! We had the opportunity to showcase the work at both IDM showcase as well as the ITP show.
(all te picture below are from IDM show on the saturday).
After trying both venue, we realize that our theme & dynamics of the work feels more align witht the mysterious empty-with-echoes- feeling from the Navy Yard Venue. The intimate setting at the ITP show is different and interesting, and we feel would perhaps be very suitable for very personal pieces.
Tech Check @MoCap Room at The Yard, IDM
This is the first&last chance we officially test the MoCap Room equipment before the IDM show.
We succesfully
- connected & tested the big LED screen, see the visual’s (made in touchdesigner) effect
- the sound, which would be controlledby the main computer
- the floor projection, which is projected by two projectors mounted on the ceiling each taking care of half of the stage/ space. (with th help from Luke)
- an OSMO camera from professor Louise at IDM, that we put it up on a tall tall tripod.
As the original visual that I made for the floor is a little bit too dark, it is not obvious enough given that there would be light sending out from the LED screen. So I re-made another verison for the audio-reactive floor projecton-- I further explored the POP and particle system in TD! was super fun!
Tech Rehearsal @220 Black Box, Media Commons
This was the first time we tested with the intended farbic into the dance -- the material is super lovely and beautiful. We adjusted the length & width of the fabric, and further polished the choregraphy.
With a few time of the run through, we flet prepared for the performance in the upcoming weekend :)
Tech Check 3rd Rehearsal
We incorporated in the physical props & projection (without the intended projection fabric) & other technical elements in this rehearsal (which took up way more time than expected -- a note for the future rehearsal & performance, for how long the installation & adjustment would be).
For the visual, we briefly tested a touchdesigner version.
from the tech check in Media Commons Black Box, we confirmed that we want to use floor projection for the light source that Kinect could detect human bodies, as well as multiple semi-transparent meshs to be the projection background.
2nd Rehearsal
Today’s rehearsal, we briefly show the code version of the visualization to the dancers -- how one person is a dot, two people becomes a line, three people becomes the triangle, etc.
Together with the dancers, we workshopped through all the ideas & have a more concrete idea on the choreography -- how the storyline go through and get reflected through the choreography, interaction, and audio-visual.
Days of Making
Before making any changes, I revisit the code and ran and logged it with the footage from Sunday. One thing worth noticing is that, the wrist detection often is pretty hard, when the dancers‘ bodies overlay or face to different direction. Sometimes the general body’s presence could be detected, but the right writst (very specific point) cannot be detected or logged.
model being used: poseDetection.movenet.modelType.MULTIPOSE_LIGHTNING
library: Tensorflow
NEXT STEP FOR TODAY:
1st Rehearsal
We had our first rehearsal with our four dancers, two street style practitioners and two classical style dancers (from NYU Tisch) for the first time— it was truly amazing to see the ideas & thoughts coming alive and together through the moving bodies and energy.
After introductions, project and story run through, we explained the mood, and listened to some sampled music together with dancers. Four dancers freestyled/ improvised one by one, to two tracks (one calmer track, and one more intense track). We decided the role according to each dancer‘s movement mood and quality and energy.
1st Iternation
I first tried the Media Pipe, which generate a result that is okay.
But the detection between two peple are constantly very glitchy -- it got super confused about who is who.. Even though the glitchiness could make its own point, for the visual consistency for this project, I want something that’s more steady. So eventually changed to using TensorFlow.
Building an upload video button for now, to be easier to test.
days of making - boundary, body
this week’s inspiration come fron the boundary/ border of the body/ range of motion— when dancing/ practicing either choreography or free style, we alwys talk about the range of motion. How large your body could expand. How much space your body is taking. When in Lisa’s Body Everywhere class, I also liked the example of S.P.A.C.E by ELEVENPLAY x Rhizomatiks.
When trying the multiple people’s version, which used the tensor flow.net (the recognition of two people is a bit glitchy), it’s really interesting to imagine what could happen when two people gets two close. An possibility brought up by Cassidy, is even that, what kind of physical world’s output could this close of distance trigger -- sound, light, etc?
Later on, I managed to use newer version of @mediapipe/tasks-vision, in which I could edit to change the max number of people, so it could detect multiple people -- for interaction between bodies.
I also tried coloring of the space-- either the space that the body contains, or the space that’s outside at th body.
days of making - hand throw text
what would happen if we could interact with the language/ word/ spoken words?
This week’s inspiraiton really comes from the idea, of me imagining two people throwing/ transferring words/ spoken words to each other.
I only got to test with one person -- the multiple people’s one seem to be a bit janky, especially the hand recognition and throwing motion
if to iterate on this one, I would love to see how two people could interact with the same camera, or even across the screen? if in a performance setting, how large the screen(s) could be, and how wide the camera could detect?
days of making - finger + text
Starting off this 30 days of making series, I experimented with the idea of letting spoken words appear on the finger tips.
The inspiration come from my own experience of watching performance, as well as seeing Bill T Jone’s collaboration with Google Creative Lab, when he was doing his classic monologue + performance.
I remember back in one of my college dance lab showcaes, one of my classmate’s solo piece -- 1 dancer in the middle, and the choreographer sitting off stage on the side, reading their own peom as the monologue of the dancer -- it was one of the best performance I’ve see in college, we are all left with tears in our eyes. The power of sincere words + emotions, and movement, without even music it’s still touches our heart.
All that to say, I’m curious to explore: what would happen when the spoken words could appear together on the stage/ screen with the dancer’s moving body? Also, the potential of using body/ finger in this case to be a visual medium in 2D space of displaying the words?
When you are expressing through spoken language, how would you want to extend the expression further through your physical body?
Something I found super intriguing, is that when two hands get close to each other, it starts to fluctuates a lot, as the model being unsure where is the left hand and where is the right hand -- the TENSION appears when the words start to flash between appearing & disappearing.
What kind of emotion could be pair so well with this fluctuation/ tension on the visual display?