Okutama-Action

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #ai #typescript #llm #go #golang #android #machine-learning #rust #deep-learning #linux

(★ 75)

Okutama-Action: An Aerial View Video Dataset for Concurrent Human Action Detection

Okutama-Action Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

#### Abstract

_We present Okutama-Action, a new video dataset for aerial view concurrent human action detection. It consists of 43 minute-long fully-annotated sequences with 12 action classes. Okutama-Action features many challenges missing in current datasets, including dynamic transition of actions, significant changes in scale and aspect ratio, abrupt camera movement, as well as multi-labeled actors. As a result, our dataset is more challenging than existing ones, and will help push the field forward to enable real-world applications._

#### Publications

[Okutama-Action: An Aerial View Video Dataset for Concurrent Human Action Detection](https://arxiv.org/abs/1706.03038)

#### Download

- Sample (one 4K video and labels): [sample](https://www.dropbox.com/scl/fo/9qvpsb3fsamvqzsa12149/APTyV-f01XLnJ0WFpZSBLOE?dl=0&e=1&preview=Sample.zip&rlkey=7u7131amaul29amyr4jbnnu03)

We offer the dataset in two different formats:

- 1280x720 frames + labels: [train set](https://www.dropbox.com/scl/fo/9qvpsb3fsamvqzsa12149/APTyV-f01XLnJ0WFpZSBLOE?dl=0&e=1&preview=TrainSetFrames.zip&rlkey=7u7131amaul29amyr4jbnnu03), [test set](https://www.dropbox.com/scl/fo/9qvpsb3fsamvqzsa12149/APTyV-f01XLnJ0WFpZSBLOE?dl=0&e=1&preview=TestSetFrames.zip&rlkey=7u7131amaul29amyr4jbnnu03)
- 4K videos + labels: [train set](https://www.dropbox.com/scl/fo/9qvpsb3fsamvqzsa12149/APTyV-f01XLnJ0WFpZSBLOE?dl=0&e=1&preview=TrainSetVideos.zip&rlkey=7u7131amaul29amyr4jbnnu03), [test set](https://www.dropbox.com/scl/fo/9qvpsb3fsamvqzsa12149/APTyV-f01XLnJ0WFpZSBLOE?dl=0&e=1&preview=TestSetVideos.zip&rlkey=7u7131amaul29amyr4jbnnu03)

In addition, we provide trained models in Caffe: [models](https://www.dropbox.com/scl/fo/9qvpsb3fsamvqzsa12149/APTyV-f01XLnJ0WFpZSBLOE?dl=0&e=1&preview=FinalModels.zip&rlkey=7u7131amaul29amyr4jbnnu03)

All files: [link](https://www.dropbox.com/scl/fo/9qvpsb3fsamvqzsa12149/APTyV-f01XLnJ0WFpZSBLOE?rlkey=7u7131amaul29amyr4jbnnu03&dl=0)

#### About

The creation of this dataset was supported by [Prendinger Lab](http://research.nii.ac.jp/~prendinger/) at the [National Institute of Informatics](http://www.nii.ac.jp/en/), Tokyo, Japan.

Okutama-Action

Okutama-Action Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Okutama-Action

Okutama-Action Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README