This week I continued working on our model using our custom data. I ran into some issues with data resizing so I worked mainly on adjusting the code so that it could work with our custom data. As of now, the output of the model is not completely satisfactory, but I will continue working on it this week. In preparation for the demo, we also mounted the mirror on the TV and tested in person.
Devon’s Weekly Status Report April 10
This week I worked mainly on training the model that we will be using for our clothing recognition and image warping. I had some difficulties training initially on my computer as CPU-only so I put everything on Google Colab and trained from there. I will experiment with different training methods once we deploy it on the Jetson. Judy and I built the TV stand and I helped her with some of the set up of Open Pose on the Xavier. I think that we will be in a good place for the demo on Wednesday and I’m excited to see everything start to come together a bit more. I feel like we are currently working at a good pace, and hopefully things will go smoothly next week.
Devon’s Status Report for March 13
This week we began to implement our torso and clothes detection algorithms. I played around with OpenPose to get comfortable with it and see how it might work with the Xavier NX. This was a relatively slow week as we had our presentation but things will definitely pick up next week!
Team Status Report for March 13
This week we presented our design presentation slides. This went well and we got some useful feedback from students and professors. We each began getting more familiar with our aspects of the project – mainly the UI and software, for now until our materials come in and we are able to start implementing with hardware. We are testing and experimenting more with OpenPose and our UI.
Devon’s Weekly Status Report for March 6
This week, I worked on the application area, solution approach and hardware block diagram slides for the design presentation. To further specify our design, I researched the current market space for smart mirrors similar to ours. I discovered that there are three main areas of smart mirrors: workout mirrors (like the MIRROR product), vanity smart mirrors (such as HiMirror, a product that gives you feedback for makeup and skincare), and DIY smart mirrors. There is also a subspace of commercial smart mirrors for clothing department stores but these are inaccessible to everyday users. This was valuable information since our product will take a unique approach in the smart mirror market by allowing the user to try on clothes. In addition to the market space, I also researched the construction and testing procedures for these products. I then looked into how the hardware components in our design will specifically function with each other. We decided that it is better to use the singular Arducam rather than the two depth-cams for our project. I also put in the orders for our parts so that we can begin assembling ASAP.
Devon’s Status Report for February 28
I researched other similar products and investigated their user spaces as well as looked into how the products were implemented. I looked further into how we will use OpenCV, OpenPose and DeepMark together on the Jetson Xavier NX. I hope that next week we will be able to get our design completely finalized so we can begin developing.
Devon’s Status Report for Feb 20
This week I worked on creating my slides for the project proposal presentation and editing our website. I researched software and hardware solutions for our project as well as brainstormed technical challenges we might face.