The problem HotelOPS.AI solves

Revolutionizing Hotel Operations with AI-powered Room Management.
This project presents a novel AI-powered solution that streamlines hotel operations and enhances guest experiences. Our model leverages image recognition to tackle three key challenges:

1)Automated Mess Detection: Eliminate time-consuming manual checks by automatically identifying messy rooms through image analysis.
2)Efficient Inventory Management: Ensure accurate and swift inventory checks using image recognition, reducing discrepancies and streamlining stock control.
3)Proactive Damage Identification: Identify broken items within rooms during the photo inspection process, allowing for prompt maintenance and replacement, minimizing inconvenience for guests.

By automating these tasks, our AI model empowers hotels to increase operational efficiency, reduce costs, and guarantee consistently clean and well-maintained rooms, ultimately leading to higher guest satisfaction.

Challenges we ran into

Challenges Faced and Solutions in HotelOPS.AI Development :-

Challenge 1: Limited Data & Model Performance: Our initial CNN model struggled with a small dataset, leading to suboptimal results.
Solution: We leveraged pre-trained models, but required fine-tuning the data generator to address the data size limitation.

Challenge 2: Inventory Detection Model Selection: YOLO-V5, initially chosen for inventory detection, presented unforeseen issues.
Solution: We successfully adopted the GEMINI-API with a Google API key for robust inventory detection.

Challenge 3: Deployment of Large Model: Deploying the hefty model using a .hy file posed a challenge.
Solution: We implemented ngrok to effectively deploy the model despite its size.