Prior to this week (Week 4), we were able to identify one of the most significant risks of the People Counter project as the complexity of our system. Originally, we were planning on only implementing and testing an estimation system that allows users to view the occupancy data of each enclosed space we monitor. However, after meeting with the professor and our TA, we realized that the project’s scope may be too narrow in terms of feature design and implementation. So, over the course of this week, we have been taking action to mitigate that risk by designing an additional prediction feature. This pivot allows our system to be more interactive, as users will be able to select a room and a time of their preference, and the system will show them the predicted occupancy (category-based) given the parameters. We believe that this feature will greatly increase the overall complexity of the system and makes it more suitable for a semester-long project. Additionally, the usability of the final product will also improve, as the UI will feature an interactive element.
Besides the aforementioned change in feature development, we designed a more detailed version of the tech stack and workflow of the project.
One of the notable changes is that we are placing a greater emphasis on accurately and promptly translating the CV output to occupancy data, which, as we see it, is one of the most crucial tasks to complete in order to meet the use case requirements outlined in our proposal presentation. While the change in our system design will not induce an immediate change in overall cost (besides time), we as a team are planning on to add a lot more details to this graph and our overall design in the coming week.