Posted on

Jun Wei’s Status Report for Apr 19 2025

1. Personal accomplishments for the week

1.1 Stitching algorithm integration

This week, I completed integration of my motorized camera system with the image stitching algorithm. The issue right now is the amount of time required for the image to successfully stitch, posing additional issues with meeting the overall pipeline time constraints.

1.2 Integration with cloud transmission pipeline

I have also begun working with Steven on integrating the motorized camera system with the cloud infrastructure he has been using for CV inference. There are some issues with getting transmitting the image to the cloud database.

2. Progress status

I am on schedule as the camera system has been integrated with the stitching algorithm. All that is left is multi-system integration.

3. Goals for the upcoming week

  • Complete integration with cloud transmission pipeline (to work with Steven on this)
  • Complete integration with the smartphone application.
Posted on

William’s Status Report for Apr 19 2025

This week, I pivoted from using LangChain to a machine learning–based approach for the recipe recommendation system. Instead of relying on prompt-based querying, I’m now representing each recipe as a feature vector derived from its ingredients and tags. Similarly, a user’s fridge contents and dietary preferences are converted into a query vector. Recommendations are generated by scoring the similarity between the query and each recipe vector, returning the top 5 most relevant matches.

In parallel, I made minor UI adjustments to improve the layout and clarity of the Preferences and Recipes tabs, helping prepare the app for future user testing.

Plans for Next Week

  • Finish implementing and testing the vector similarity scoring system across a sample recipe dataset.

  • Integrate the recommendation output with the mobile UI.

  • Explore options for caching or pre-processing recipe vectors for performance optimization.

This new approach gives us more control and flexibility over how recommendations are generated, and sets the stage for fine-tuning the system as we scale.

Posted on

Steven’s Status Report for Apr 19 2025

For this week, I deployed our YOLOv5 model onto the cloud and used it to run inference on fridge images. Furthermore, I refined our data upload pipeline, though there are some intermittent failures in the cloud upload service which requires more reliable error handling and try mechanisms.

Currently, we are on track and are finalizing the integration of our model with the Raspberry Pi and the cloud pipeline.

For next week, we will be conducting our final presentation. I will work to further improve cloud pipeline reliability and finalize integration between our components for our final demo.

Posted on

Team Status Report for Apr 12 2025

Team Status Report

Progress This Week

Our team made meaningful progress across hardware, cloud integration, and software components in preparation for the next stage of development.

  • Computer Vision & Cloud Integration: Steven worked on integrating the CV model and cloud components of the system. He successfully tested image upload and retrieval through the cloud pipeline and began preliminary integration of the YOLOv5 model for live inference. Training of the model is ongoing, and initial tests on the cloud system have been promising.

  • Hardware & Imaging: Jun Wei completed the construction and integration of the motorized camera slider with the updated camera hardware. The system has been mounted onto a smaller breadboard and tested within the mini fridge enclosure. He is now moving toward integrating the image stitching script with the Raspberry Pi system and preparing for cloud pipeline integration in collaboration with Steven.

  • Mobile Application & Recommendation System: Will implemented Google OAuth in the mobile app, enabling secure user sign-in and laying the groundwork for personalized features. He also started building the recipe recommendation system using LangChain, setting up the initial framework and exploring prompt designs to map fridge contents to relevant recipes.

Plans for Next Week

  • Finalize deployment of the YOLOv5 model for cloud inference and continue refining the cloud data pipeline (Steven).

  • Integrate image stitching into the Raspberry Pi workflow and begin transmission testing (Jun Wei).

  • Continue development of the recipe recommendation system and begin linking it with user preferences stored post-login (Will).

Overall, the team is on track with project milestones. Core components across hardware, CV, and app infrastructure are coming together, and collaboration between sub-teams is increasing as we begin cross-component integration.

Posted on

Jun Wei’s Status Report for Apr 12 2025

1. Personal accomplishments for the week

1.1 Motorized camera system integration

This week, I completed construction of the motorized camera slider and integration with the new camera obtained. I found a way to tether the movement of the timing belt to the camera baseplate using zip ties. I have also managed to move the system to a smaller breadboard and have tried placing it within the mini fridge that we will be using.

2. Progress status

I am on schedule as the motorized camera slider has been integrated with the camera system. I am now working on integrating the stitching algorithm script with the overall Raspberry Pi script.

3. Goals for the upcoming week

  • Integrate stitching algorithm
  • Integrate with cloud transmission pipeline (to work with Steven on this)
Posted on

William’s Status Report for Apr 12 2025

This week, I implemented Google OAuth integration into the mobile application, allowing users to sign in seamlessly and securely. This sets the foundation for saving personalized preferences tied to user accounts.

I also began working on the recipe recommendation system using LangChain. Initial efforts were focused on setting up the pipeline and experimenting with different prompt structures to connect user inputs with relevant recipe suggestions.

Plans for Next Week

  • Finish building out the recipe recommendation system and fine-tuning its performance.

  • Start linking user preferences to recipe suggestions post-login.

  • Explore potential improvements to the UI for the Preferences and Recipes tabs.

Progress this week has been steady, with key backend components starting to come together alongside the app frontend.

Posted on

Steven’s Status Report for Apr 12 2025

For this week, I worked on integrating the CV and cloud components of our system. I tested image upload and retrieval through the cloud data pipeline, as well as preliminary integration of the YOLOv5 model onto the cloud system for live inference testing.  I’ve also continued further training of our YOLOv5 model.

I am currently on track with our milestones. I am working on integrating our model with our cloud pipeline which will also be integrated with the Raspberry Pi.

For next week, I plan to finalize the deployment of the YOLOv5 model onto the cloud and run cloud inference on fridge images. Furthermore, I plan to continue refining the data upload pipeline and explore backup storage solutions in the case of a cloud failure.

Posted on

Steven’s Status Report for Mar 23 2025

For this week, I continued with refining the training pipeline for the object detection model, in order to maximize our performance for the demo . I expanded our dataset with the Open Images database, filtering out food-related classes to our use case, significantly increasing our existing data. Furthermore, I am looking to experiment with the YOLOv10 model in order to further increase accuracy.

I am currently on track with our milestones. Preliminary training have been completed, and we are working to maximize our accuracy and finalize our model. I am also working on deploying our model onto the Raspberry Pi.

For next week, we aim to proceed with our demo, during which I aim to show our model with the optimal accuracy. I will continue working on integrating our model with the tentative pipeline for fridge item detection, and continue expanding our training dataset and tuning our parameters to optimize accuracy.

Posted on

Steven’s Status Report for Mar 16 2025

For this week, I continued with developing my core image processing pipeline. I conducted tests on object classification using the YOLOv5x model with some sample fridge images, and I collaborated with my team to finalize the list of our target food items for model training.  I always begun with integrating the CV pipeline with the Raspberry Pi to validate real-time image capture.

For my progress, I am currently on track with my project timeline. Preliminary object detection is functioning, and I continuing to work on improving the model frequency.  I am beginning on integrating my model with the edge processor.

For my goals for next week, I aim to continue improving the performance of my YOLOv5 model, as well as integrating the model with our Raspberry Pi.

Posted on

Steven’s Status Report for Mar 9 2025

For this week, I continued with data collection, expanding our datasets by importing annotated images which I have found online. I have also performed further fine-tuning of our training hyperparameters, testing different augmentations and hyperparameter optimizations in order to improve our detection accuracy, incorporating the expanded dataset in order to improve the training process. I then ran a preliminary test for our YOLOv5x model to test its detection accuracy. As expected, the YOLOv5x model performs at a higher accuracy, though it results in a longer inference time when run locally.

In term of progress, I am on track by improving model accuracy while also maintaining real-time performance.  The next step in our progress chart is integration, and I will begin work with integrating our model with the Raspberry Pi and the model app.

For next week, I aim to run inference on cloud and measure inference timing in order to quantitatively assess the latency of our model. I will also continue to work on training and enhancing our YOLOv5x model, and conduct more real-world tests to measure the performance of our optimized model.