Traversing Spacial Perspective

     

        My project is a traversing experience through the 3d version of a scroll painting Spring Morning in the Han Palace.(See image above)Focusing on the study of the experimental perspective technique – the cavalier perspective – use in traditional Chinese painting, the project studies the contrast and connection between the traditional western drawing perspective technique and the eastern drawing perspective technique.

 

  • Technique&Workflow:  

The first idea of me is to “photogrammetry just through one picture”

Runway ML for depth estimation + surface displacement in blender

ML Code Credit:GitHub – isl-org/MiDaS: Code for robust monocular depth estimation described in “Ranftl et. al., Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer, TPAMI 2022”

        The technique of depth estimation can generate a general gradient of the grey-scaled image from a single picture. The gradient logic is: the white part represents the components that are closer to the camera, and the dark part represents the components that are farther away from the camera. Then I applied the gradient map as a displacement map to the subdivided surface in blender- it will render a “bumpy” surface and created a semi-3D model.

  • Test Cases:  
      • Test Case 1: Coffee Cup

Actual Picture        &        Depth Estimation Generated Gradient Map

Blender Model

 

        For further study – instead of actual photographs – I choose to process oil paintings, intending to bring out 3D spaces from unreal worlds. 

 

      • Test Case 2&3: Edward Hopper & Dali
          • unreal content

2D Oil Painting       &        Depth Estimation Generated Gradient Map

 

2D Oil Painting       &        Depth Estimation Generated Gradient Map

And I put it in Unity!

        These two models work well because they follow the rule of the classic western drawing technique: perspective. (either 1-point perspective or 2-point perspective that required vanishing points)

However, in ancient Chinese scroll paintings, the painters do not apply point perspective as a painting technique. They instead use cavalier perspective – also called scattered perspective – in their long scroll paintings.

Cavalier perspective(scattered perspective)  & Oblique Elevation perspective

“Oblique Projection is a type of technical drawing that is used for drawing graphical projections of three-dimensional objects on two-dimensional planes. The most common technique used for technical drawing is Oblique Projection.

From the first or second century to 18th-century Oblique Projection was used almost universally by Chinese artists for depicting rectilinear objects such as houses.” 

cr. What Is Oblique Drawing | Oblique Drawing Examples | What Is Oblique View | Oblique Projection | Oblique Shape | Cabinet Oblique | What Is Cavalier Drawing (civiljungle.com)

 

  • Test Case 4:  Ying Qiu – Spring Morning in Han Palace (1540—1544 AC)
    • unreal perspective!

“With every step I take, the view is also changing”

-Ying Qiu -漢宫春晓图 Spring Morning in Han Palace-

 

  • Final Model:

 

 

 

 

 

 

 

 

 

 

 

 

Final Project – Across Dimensions

What: I’m going to use depth analysis to create a “3D version” of 2D printings – which can create a different perspective to experience the paintings in the museums.

Technique: Depth Analysis + Blender

depth analysis can create a really interesting outcome with perspective drawings.

– Test 1

Test2: Edward Hopper

These are some rough drafts I created and tested. One problem is that they are somehow ragged, and might need further editing with the surface details.

I vision the final project could be 1. a combination with AR techniques that people can dynamically viewing the model from multiple perspectives or 2. Transport the blender to unity for 360 experience.

 

 

Human In Time – The Silhouette

Interested in Bauhaus’ study of geometry shapes done by Oskar Schlemmer, I want to study more about the relationship between abstract geometry and the human body.

This project is a person dancing,  wearing different forms of geometry on their body, improvising to a piece of music, and trying to express their feeling with the shapes they are wearing.

I played with the frame rate and the transparency of the video .

 

Challenges:

  1. The silhouette of the human body can distract people’s attention on the geometry shapes.
  2. The original videos are too large for my computer to process so only a few short clips are successfully edited 🙁

Proposal

Nu descendant un escalier n°2

I want to explore more geometric forms that a person can play with. Following the logic of Oskar Schlemmer’s study of the living sculpture, I want to let the same person wear different geometric shapes on their body and then record their body’s movement and geometric shapes.

(I have a friend doing contemporary dance, so I want to record their improvising on the same piece of music for this study.)

geometric shapes can come from Kandinsky:

LEAF AS MAP – LEAF SKELETONS TYPOLOGY

 

This project is a typology of the extraction of leaf skeletons.

    As someone who has always been fascinated by the systematic and complex nature of maps, I somehow found that the skeleton of leaves speaks a similar language to maps. They are all systematic, well-structured, and direct a flow of nature. 

 

 

    In the past few weeks, I’ve collected 70+ leaves from North Oakland, Shadyside, and Squirrel Hill areas as a flaneur to start my project. After that, I put leaves on a LED light board and documented them with a camera to catch more details of them. Then I use photoshop, and imageJ to read the binary information in the image to extract leaves skeletons. I also collected a lot of city maps: sometimes the city maps fit so well into the leaf’s skeleton, and sometimes the city maps show some similarity with the leaf’s structure.

   Throughout the process of extracting the skeleton of leaves, the connection between the human-made structured map and the nature-made leaf skeletons is gradually shown. Compared to conventional maps, Leaf skeletons are more organic and invite people to think: where are these branches leading? Does every piece of leaf serve as a mini portrait of the great place where people lived in?

   In general, I’m very fascinated by the details that a leaves skeleton has given me and I enjoyed the process of collecting leaves and observing them, which is like a practice of meditation. Sometimes while extracting binary information and skeletonizing the images, some details will surprisingly come out and create a pattern that I cannot read with human eyes.

    I also learned how to simplify an art idea and do it well. Extracting leaf skeletons is not as easy as I thought: It indeed needs very careful documentation and processing.

    One thing that needs to be put more effort into is the extraction process. I did not make a clean “skeletonized” diagram because the textures and patterns of different leaves are very hard to control, so there were some problems that occurred when I tried to use the “make binary” command: some detailed information was lost and thus lost consistency of the leaves. Due to this reason, even though I’ve collected 70+ leaves, it turns out that only 20+ leaves are successfully extracted.

Leaves as map

My project is gonna be “Leaves as map”. The skeleton of leaves somehow speaks the same language as maps to me. The skeleton of leaves are random but organized patterns seem like they can lead people somewhere.

To represent the connection and similarity between the system of leaf skeleton and man-built map, the first thing I’m gonna do is to scan the skeleton of a set of leaves(maybe three)

Then I’m going to set up three groups of 5 people to aim for their assigned direction starting from the same point (point state park maybe) and use a GPS app to track them. People are not going to have clear instructions on how they are going to get there but using their own will to go to the destination as a flaneur using a compass.

At last, I will overlap the GPS pattern of the path that each people has walked and the leaf skeleton to expose the similarities and differences between how people construct a path and how mother nature forms a path. This process of meditating trip is the “machine” that I’m going to make, and the final typology outcome will be a set of diagrams as shown in my drawing.

Shrimp tail SEM scanned pics

The object I scanned is a shrimp tail(cooked cuz I took it from my thai fried rice). The first pic is the feathers that shrimp used for sensing water flows but with grease on it. The second pic i found it so amazing is that it is the broken joints of the feathers on the tail and it even looks like animal skeleton(or teeth).

Read-1 Reflection

  As a person who has taken photography as an artistic expression for a long time, I now find out photography originated from the scientific observation of the surrounding world. It originated from scientific documentation(which I somehow always ignored) to a more sensational art expression. Also, It is very exciting for me to explore the limitation of photography actually. It started from how people differentiated on their own biases, background, and perspectives. Photography is limited because “it only captures a specific object at a specific time in a specific place.”(17.) It is fascinating to know because people would have endless perspectives to observe this specific object. Other than photography, photogrammetry is a more analytical form of representation. By using mathematical, econometric, or other scientific knowledge with the combination of playful shooting skills, we can observe images that the human eye cannot perceive: for example astronomy or anatomy area.

  I see the artistic opportunity in quantifying everything happening around us with scientific methods. From my perspective, the information in the world is overloaded, and photography is a way to extract, organize, and re-represent the information I obtained. This re-representing process would give out images that people would ignore. Using the scientific method, I can organize the information by time or projectile and then juxtapose them together and build a connection between them. In the future study process, I would explore more using scientific techniques to extract more organized information, and explore more photogrammetry.

PARIS | Guy Trefler

PARIS | Guy Trefler

Work title:Paris

Author:Guy Trefler

Type: Short Video

This video shows a mixing of reality and personal interpretation and his own imagination. He alternates between the real urban composition in Paris and the wild, crazy, also childish doodles in this video. I think what inspires me in this video is what the camera already captured can also be changed! By doing this, Guy said that he created a new story for Paris and I think it gives people multiple perspectives to interpret the city.