PoolTracker DDR

This week most of the progress that I made was directly with the group. I worked with Jack, on fine-tuning the Tracking algorithm by incorporating some preprocessing steps that would improve the reliability of our tracker. Furthermore, I worked on creating the interface between the front end and the back end by writing some of the url bindings for our web app. Currently, we are able to detect wall touches and send + visualize the time data on our web-app.

For the upcoming week, our focus is to have a fully functional demo except for stroke classification (for which we have some code set up as well).

This week, most of my progress has been made working with the group as a whole. Because we are putting together our demo, it’s important that we’re meeting as a team to synthesize the different components of our project.

Adithya and I worked together to make the swimmer tracking much more reliable than it was before. We are now able to do wall detection, which allows us to calculate lap times and send it over to the web app. I also worked to put together the graphing capabilities on the front end, although there is a lot of growth here from a UI standpoint.

In the next week, I’ll be focusing on polishing things up for the Wednesday demo. After that, I’ll be refocusing my attention towards another one of my classes, where I’m helping to build a dark ride for Carnival. With load-in beginning on Friday, I do not expect to have much time for capstone next weekend.

This week, our team has been putting together the different components of our project for the upcoming demo. In particular, we have focused on getting the backend to communicate with the frontend and update the web app every time a swimmer completes a lap.

Currently the biggest risk to our project is stroke classification, so we have one team member dedicated just to setting up an infrastructure for this. Even though stroke classification won’t be part of our demo, we decided to collect additional aerial footage that we believe will be more conducive to this task.

All of the time that I spent on the project this week was working with the group. Because a majority of our accomplishments this week required communication between the different components, this is what made the most sense. I focused mostly on changes with the html and javascript that were required for our changed to the UI.

One of the most important strides was getting the web app to automatically refresh with whatever new data had been sent from the backend. This involved using Ajax to regularly check for updates to graph.

In this coming week, I will be focusing primarily on the work flow of the user with our web app.

This week, the majority of the progress came on the web app side of things. We designed the overall UI of our web app and at the same time got it to respond to requests from the processor back end.

Additionally, we started to spend some time with different tracking algorithms in order to check what works well and what doesn’t work well for our particular problem domain. We have a couple interesting solutions to solve the problems that we face.

By the upcoming demo in 2 weeks, we plan to get our tracking completely working so that the only thing that we have as a remainder after carnival is the stroke classification task as well as minor tweaks to our project.

This week almost all of the time that I spent on this project was spent on working on the same problems as the rest of my group. As a result, I worked on implementing the web app and at the same time getting some tracking algorithms tested.

For the upcoming two week, my personal aim is to get tracking working completely. Additionally, I will continue to assist my group with the web app.

This week, there was very limited time to work on our capstone projects due to other commitments, such as midterms. However, we did see some progress from the last few weeks. First, our group managed to set up limbic detection with OpenCV, which serves a key part of our project in stroke classification for pools. Additionally, we managed to segment the video via lanelines, again using OpenCV, and track each swimmer in each lane individually. Finally, we have not yet faced issues with the web-app component of our project, which is nice.

In short, given the recent developments in our work, I’m optimistic that we will have a feasible project up and running by the end of the semester.

As expected, there was limited time to work on the actual project this week due to midterm deadlines as well as the shortened week for Spring Break. Writing the design review report took a lot of time, as we realized there were aspects of the project we still didn’t know we could accomplish.

One of the most important realizations we had is that stroke classification is a much more difficult task than we originally thought. Adithya got OpenPose running on his laptop, but it was unable to perform limb detection on any swimmer. Without limb detection, we can’t even begin to extract information that would be meaningful to a classifier. Because of this challenge, I began searching for alternate ways we might be able to classify strokes without limb detection. The most promising of which is to use k-means clustering to differentiate between the swimmer’s body and other parts of the pool. Upon initial testing, we appear to get the best results with either k=2 or k=3.

I believe that some pre-processing could drastically improve these results, and I’ll be focusing on this in the coming week. I will also be creating the overall UI of our web app since I was supposed to have that completed this week.