Sign in to view source links and access this dataset
Description
8157 anime and art images, including manga pages, annotated with object bounding boxes. The dataset is split into training, validation, and testing sets and was created by nyuuzyou. It was last updated on July 2, 2023.
Use Cases
Train object detection models based on annotated anime and manga images.
Benchmark model performance on a dedicated anime art dataset.
Fine-tune pre-trained detectors for character or object recognition in illustrated media.
Strengths
Contains 8157 annotated images, providing a substantial base for model training.
Includes a standard split into training, validation, and testing sets for structured evaluation.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Last updated 2023-07-02 23:24:38; freshness should be verified.
Data may reflect source bias inherent to the collection of images from various unspecified sources.
Provenance
Source
nyuuzyou
Collection Method
Images were collected from various unspecified sources.
Freshness
Last updated 2023-07-02 23:24:38
License is unknown; terms of use must be verified before application.