VOC2007Augs / README.md
danulkin's picture
Create README.md
f34855b verified
metadata
license: mit
task_categories:
  - object-detection
language:
  - en
size_categories:
  - 1K<n<10K

Dataset Card for Dataset VOC2007Augs

Dataset Details

Dataset Description

VOC2007Augs is a dataset containing augmented images from the VOC2007 dataset, specifically focused on two object classes: cats and potted plants. The dataset is structured in the VOC2007 format, with two separate folders, CatAugs and PottedPlantAugs, each containing a VOC2007-style directory structure.

Uses

This dataset is suitable for use in object detection and image classification tasks, particularly those focused on the cat and potted plant object classes. The augmented images can be used to improve model robustness and performance on these tasks.

Dataset Structure

VOC2007Augs
β”œβ”€β”€ CatAugs
β”‚   β”œβ”€β”€ Annotations
β”‚   β”œβ”€β”€ JPEGImages
β”‚   β”œβ”€β”€ ImageSets
β”‚   └── Prompts
└──  PottedPlantAugs
    β”œβ”€β”€ Annotations
    β”œβ”€β”€ JPEGImages
    β”œβ”€β”€ ImageSets
    └── Prompts

Dataset Contents:

The dataset contains images, split between the CatAugs and PottedPlantAugs folders. Each folder contains a VOC2007-style directory structure, with the following contents:

  • Annotations: XML annotation files for each image, containing bounding box and object class information.
  • JPEGImages: The augmented images themselves, in JPEG format.
  • ImageSets: Text files containing lists of image IDs for training, validation, and testing splits.
  • Prompts: Text files containing the prompts used to generate each augmented image.

Dataset Format:

The dataset is stored in the VOC2007 format, with XML annotation files and JPEG images.