Meta-Album ACT 40: Stanford 40 Actions Image Dataset

800 images of humans performing 40 distinct actions, preprocessed to a uniform 128x128 pixel resolution. The dataset is a curated subset of the original Stanford 40 Actions dataset, released under the Meta-Album project for few-shot learning. It was created by Jilin He for Meta-Album in March 2022.

Use Cases

• Benchmarking few-shot learning algorithms based on the 40 distinct human action classes.
• Training models for human action recognition in still images based on the described image content.
• Developing lightweight computer vision models that leverage the standardized 128x128 pixel image size.

Strengths

• Contains 800 images across 20 classes, providing a structured benchmark for few-shot learning.
• All images are preprocessed to a consistent 128x128 pixel size, simplifying model input pipelines.
• Clear provenance linking to the original Stanford 40 Actions dataset and associated research paper.

Limitations

• Dataset scale is limited to 800 images, which may not be sufficient for training large-scale models from scratch.
• Column-level documentation and sample data are unavailable, requiring inspection after download to understand data structure.
• The non-commercial (CC BY-NC 4.0) license restricts commercial use cases.

Provenance

Source

Stanford 40 Actions dataset, preprocessed and released by the Meta-Album project.

Collection Method

Images were resized to 128x128 pixels using OpenCV as part of the Meta-Album preprocessing pipeline.