Tea Time Dataset

Dataset Composition:

• Images: The dataset consists of a wide variety of JPG images in different dimensions, ensuring variability in object appearance, lighting, and background. This variability helps improve the robustness of AI models trained on the dataset.

Annotation File:

• teatime_annotated.csv: This file is provided as a supportive resource to accompany the images. It includes:
  1. file_name: The unique image file name (string).
  2. teacup_count: An integer representing the number of teacups present in each image.
  3. kettle_or_teapot: A categorical label indicating whether a kettle or teapot is visible in the image.

Key Features:

• Diverse Scenes: The images are captured in various settings, from casual home environments to elaborate tea ceremonies, offering a rich set of scenarios.
• Object Variability: Teacups, kettles, and teapots come in different styles, sizes, and orientations to mimic real-world diversity.
• Supports Fine-Grained Classification: With labels for both teacup count and type of item (kettle/teapot), the dataset supports not only basic image recognition but also more detailed classification tasks.
• Scalability: The dataset is scalable and can be used for training AI models in applications ranging from smart kitchen assistants to e-commerce product recognition.