In the reading, combining Posenet and dancing can trigger words and control audio. Since there are various kinds of dancing from different cultures, I’m wondering: would there be any potential that Posenet will be able to identify dancing styles and generate corresponding music in the future? While reading the biography, I was curious how Bill became a distinct and unique collaborator with Posenet. Based on the existing technology, how would Bill promote the advancement of multi-media arts, performing arts, and AI? We learned that technology can now serve as a supporting tool to achieve our ideas and artistic expression, but I would suggest the biography look to the future and explain more potential of Bill and Posenet’s collaboration in detail.
Understanding the model and its data is fundamental and can facilitate later program development. The reading has described the process of prototype development related to human motion and AI. Throughout a loop of creating, testing, and improving projects, the team can be able to refine the model based on how Posenet interacts with the dancer's movements. Also, the article mentions that PoseNet recognizes certain points (such as the human nose) better than other parts of the body like wrists. This understanding can guide the choreographer to design certain movements or poses that the model can accurately capture. Understanding PoseNet's model allows choreographers and dancers to easily form the movements they can do based on the model's shape, in order to be better recognized by the machine.
For this assignment I changed the cat and dog & music posenet example into a drum set. Each drum on the screen has a unique sound and user would use their nose to hit the drums and hear sounds.