Show RF: Judge a Book! A POC for Book Publishers
Describe your project here! (delete this when you post)
- Link to application: https://judge-a-book.netlify.app/
- Link to project on Universe: https://universe.roboflow.com/franklin-cappadora-solutions/book-judge
- Link to product spec: Judge a Book Product Spec - Google Docs
Project Description
Judging a book by its cover isn’t a good idea, but we all do it! And publishers are well aware.
Book covers need to capture the essence of the book as well as the attention of its target audience. They can make or break the success of a book. If publishers could tap into the natural pattern recognition of a browsing shopper, they might be able to optimize the chances of someone picking it up and buying it.
This project seeks to demonstrate how computer vision can be used to find similarities between book cover appearances as a means of predicting likability. The front-end is built on Svelte-Kit using TailwindCSS and DaisyUI. It is hosted on Netlify.
How Roboflow Was Used
For MVP, a single model will be used to predict if the user will like or dislike a book cover. To train this model, a collection of book covers, each manually labeled with a subjective preference classification of “like” or “dislike”, will be supplied. The probability returned to the client with the prediction data will further quality book covers as ones they will “love” or “hate”.
As a fast follow, a second model will be trained to identify book covers from images with varying other images and settings. It will also be trained to identify when an image contains no book covers. Manual annotation will be performed using Roboflow’s built in features. Datasets from Open Images will also be evaluated. This model will not encroach upon the concern of classification, there the two models will execute in sequence using Roboflow’s workflow builder.
Going Further
The model could be expanded upon to incorporate additional metadata like genre and themes, specific sales performance metrics, reader demographics, etc., in order to further guide the design team on capturing both the essence of the book as well as the attention of its target audience.
