AI Image discriminator
The objective of the project was to develop an AI-based image filtering system, specifically trained to identify and categorize room photographs for a real estate marketplace application. The filtering system was required to distinguish pictures of rooms in a house from those that were not. To achieve this, an AI discriminator was designed and trained to effectively separate real estate-related images from irrelevant ones, ensuring that only appropriate property photos were displayed in the application. This involved training the model on a diverse dataset of real estate images to recognize various room types and features, while also developing the ability to reject images unrelated to real estate. The result was an improved user experience in the marketplace, with accurate, relevant, and high-quality visual content consistently presented.
Ask on WhatsApp