Inspiration
We were inspired by an NUS mod that we took last semester, known as IS1108: Digital Ethics and Data Privacy, and we had to do a product pitch at the end of the semester. Our product was a browser extension that reviews & verifies a web page's information by cross-checking various articles for the information stated. When we saw our problem statement, we knew that this was the chance for us to try building a browser extension to help shoppers shop more sustainably, given that we too, are chronic online shoppers who are very guilty of ordering items without knowing how they are made and their impact on the environment. We hope to empower consumers to make decisions with readily available information on sustainability.
What it does
Our browser extension will check if the current website you are on is in our database (We have taken data from The Good Shopping Guide link) and if that company has been reviewed by The Good Shopping Guide, their GSG Ethical Score will be reflected in the browser extension, and a button to learn more about the rating will be provided. While browsing products, customers can activate a reverse image search to find alternative products that look similar (through reverse image search). By clicking on the images provided, they will be redirected to the product page in a new tab. Users will also be able to provide their own reviews on brands so that they can contribute to the sustainability community.
How we built it
We used CSS and HTML for styling to give the google extension an aesthetic display, and JavaScript to make it interactive and dynamic. We used firebase to store the data of The Good Shopping Guide's ratings on brands. Then we set up a google extension which reads the current website's URL and checks back with the database on whether there is a match. When a user clicks into a product page, they can right click on the image and click on the content menu to start our reverse search image feature. The vector embeddings of the image selected will be generated by CLIP, and cross referenced to our Pinecone database. The top three images with the most similar vector embeddings will be shown on the extension popup via JavaScript and HTML.
Challenges we ran into
We found it difficult to find a good source of information on sustainability ratings, and one where we could retrieve all the data through an API. Unfortunately we did not have the finances or time to request for an API connection with some websites . So instead, we built our own database, taking information manually, and limiting our information to "Fashion Retailers" only. Furthermore, we had very little experience with software development, so we had to learn a lot on the fly by watching youtube videos or googling.
Accomplishments that we're proud of
We're proud of having a reverse image search that recommends products that look similar, and we are also proud of having GSG Ethical Scores show up depending on which brand website you pull up. Lastly, we are proud of how our extension looks.
What we learned
We learned how to build our first browser extension and how to connect the frontend to the backend. We also learnt how to build our own reverse image search which was really cool.
What's next for GreenLens
We hope to further expand the extension's reach by incorporating more brand's data into the database. We hope to add a feature where we can view the reviews submitted by users on the brand as the 'reviews' button does not do anything.
Log in or sign up for Devpost to join the conversation.