Tench Cholnoky
Projects
p.01
p.02
p.03
p.04
Bodies in H.264 is an exploration of the formal elements of digital media and H.264 encoding as well as the way human bodies and motion are encoded and decoded in H.264. By taking generic scenes from pedestrians and cars moving around public spaces in New York City, my hope was to illuminate the ways bodies are perceived in digital spaces, and the ways that video data can be creatively destroyed to learn more about its inner workings.
- Movement 1 (Glide): duplicates 6 frames and show it as a flow before reaching the I-frame
- Movement 2 (Void): clean output with distortion of pixel changes in i-frames
- Movement 3 (Noise): Adds noise to the predictions in p-frames
- Movement 4 (Reverse): Reverses the frame order, keeping p-frames and i-frames intact
- Movement 5 (Buffer): creates ring buffers around i-frames to distort predictions in p-frames
Bodies in H.264 attempts to bring the question of lossy compression to the forefront. The videos in movements one, two, and five all show intentional deletion of frames that provide vital data to the video reader. Because of the predictive frame system that H.264 compression is based on, very marginal errors in one or two frames can lead to cascading distortion throughout the video. Manipulating both the predictive frames and the key frames allowed me to better understand the relationship between them and their individual impacts on a digital video compressed in H.264.