More selected projects

Project about time and noise

The voice is the most abstract expression of time and space. The aesthetic can only imagine through the voice, and it is also the most close to the pulse and essence of life. Soundness is the general nature of all arts. The reason lies in this.

produced by: Yuwen Huang


When we were isolated at home during the outbreak, our perception of time suddenly became weak. We all know that all particles and matter have the attribute of absolute motion, because their motion, together with the various foundations such as relativity and classical mechanics, has a rule to follow in the macro motion. For example, stars move, water flows and leaves fall, people perceive these regular movements, and then they have to describe them. At this time, we introduced the concept of "time". From the scope of social science, back to the individual life, each of us has different perception of time. The world sprinter may care about 0.1 seconds, and many of us are not that sensitive. If everyone doesn't live and work from the perspective of social humanity, but makes decisions from the perspective of natural science, the degree of freedom will be unprecedented, so the world will be chaotic. In such a situation, everyone has their own time scale, and there is no recognized time. Time only exists in each individual's consciousness, that is, it is not an objective fact. When people don't exist, people's consciousness is gone, and time is gone. The second law of thermodynamics. The irreversible visible changes make people realize that something is passing. We call it time. It's hard to answer whether the concept of existence depends on human consciousness and whether it still exists without subject. But it is certain that without the subject, the concept has no meaning of existence. The trajectory of a particle moving from a to B and ab is time. If everyone stops and all the particles in the universe enter an absolute static state, does time stop.

Time is the change of one point; space is the contrast of two points (or more). Sound is the orderly flow of pitch and the harmonious contrast of this flow. The former is the time relationship, and the latter is the space relationship. Chinese traditional voice has a sense of melody, that is, a stronger sense of time; western classical voice, especially Symphony, has a clear sense of space. As for what is "order" and what is "harmony", I'm afraid we need to look for it from biological memory and social (historical) memory, and ultimately in nature. Sounds like this, what about other art forms? Painting, sculpture and architecture are space art, not time related. But at the same time when the classic moment solidifies, do great paintings often have some sense of flow, that is, sense of time? Drama, action, film and literature are also spatiotemporal arts, but how are they different from the spatiotemporal forms of sound. The time and space of action and film are concrete, which can be directly felt by the audience. The time and space of literature is abstract, which requires readers to imagine through words. The more abstract the expression, the more it can cause infinite illusion in the mind of the aesthete, and the closer it is to the essence of freedom or beauty. Poetry is the Pearl of literature crown. Impressionism has such a high position in painting. Film is more artistic than TV. Is that the reason? The voice is the most abstract expression of time and space. The aesthetic can only imagine through the voice, and it is also the most close to the pulse and essence of life. Soundness is the general nature of all arts. The reason lies in this.

So I made a program that the blue dot changes with the movement of people, and the background is a linear time orbit. The sound changes due to the motion captured in the camera, just like the feedback received by the behavior, proving the existence of time.


I used audio and color capture. The program can interact with people through the camera. The key method to identify the hand or face is color recognition. It uses a dataset recognized by a human image and marks it with a box. This enables users to clearly prompt and distinguish between face and environment. Click the face or background of the screen with the mouse to select the color. The algorithm will capture the closest color pixels and follow the motion. And 20 points are set to follow the recognition pixel and have a delay effect. The data analysis technology used is ofxOpenCv, ofxGui. The software used is openframeworks. Call the dataset Haar cascade _ frontalface _default.xml in openframeworks audiooutputexample.

Future development

If I have more time, I will add Arduino component to replace audio function controlled by color recognition. By adding Haar_like features, some features of the face can be simply described by rectangular features. Add a 3D model to replace the box after human image recognition. The closestcolor variable is mapped to the motion of the background. The volume changes when the ball touches the color block.