Week9

Group Work / Idea Changes

After the research phase of the target audience, the design of the individual project has been started and building the prototype. Before confirming and lock in the concept and I have discussed with my team member about my original idea which user finish a simple puzzle outside of the elevator and change the screen of the flat monitor inside of the elevator. However, during decision and a user acceptance test both feedbacks are the interaction are cause too long time even the puzzle are easy or not. For example, a situation if the user is doing the puzzle and the elevator has arrived on the floor, the user might just give up and go into the elevator. Therefore the concept might not work in the actual world. This reflects me that I did;t not considered enough the actual experience like time issue or social awareness, in this case, I only do what I like to do and this is not proper project development.

Therefore, I have changed my concept into the pose identification to change the screen inside the elevator as it uses less time to operate this interaction and there is common technology between other members works.

Prototype Building

After lockin the concept I have been looking for how can I code the pose identification feature to use in my prototype which I found out there is a pre-train mode call posenet that can be used to estimate the pose of a person in an image or video by estimating where key body joints are. By estimating the key body joints xy asix we can compare it and use it to pose matching to activate something.

This week still working on coding and hopefully, there will be a prototype next week which can use on testing to evaluate it.

Website of the tensor flow about the posenet https://www.tensorflow.org/lite/models/pose_estimation/overview