AVA-Dataset

KO | EN

gitlite — search

#python #react #ai #machine-learning #rust #deep-learning #nodejs #macos #java #windows #cli #kubernetes #mcp #open-source

(★ 9)

A Video Dataset of Atomic Visual Action， Google

AVA-Dataset 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# AVA-Dataset

The [AVA dataset](https://research.google.com/ava/) contains 192 videos split into 154 training and 38 test videos. Each video has 15 minutes annotated in 3 second intervals, resulting in 300 annotated segments. These annotations are specified by two CSV files: ava_train_v1.0.csv and ava_test_v1.0.csv.

Each row contains an annotation for one person performing an action in an interval, where that annotation is associated with the middle frame. Different persons and multiple action labels are described in separate rows.

#### File:
##### ./ava_train_v1.0.csv & ./ava_test_v1.0.csv
The csv files are the label set included eight keys:
- video_id: YouTube identifier
- middle_frame_timestamp: in seconds from the start of the YouTube.
- person_box(x1, y1, x2, y2): top-left (x1, y1) and bottom-right (x2,y2) normalized with respect to frame size, where (0.0, 0.0) corresponds to the top left, and (1.0, 1.0) corresponds to bottom right.
- action_id: identifier of an action class, see ava_action_list_v1.0.pbtxt
- status: the video is availabe

##### ./ava_action_list_v1.0.pbtxt
example:
```
label{
    name: "bend/bow (at the waist)"label_id: 1label_type: PERSON_MOVEMENT
}label{
    name: "crawl"label_id: 2label_type: PERSON_MOVEMENT
}label{
    name: "crouch/kneel"label_id: 3label_type: PERSON_MOVEMENT
}
```
#### Download

link:https://pan.baidu.com/s/1i4UKnqt pass:rmkw

code: mail to sibofeng AT mail.bnu.edu.cn

#### Not available video list

train set:

| video_id |
| -------- |
| 2XeFK-DTSZk |
| 4trIFq61-lk |
| EQZWzLyx-GM |
| F_-zE1dQsso |
| G3nRbyu0gMs |
| K--hW14uzA0 |
| ZFQ3lF6yq_E |
| _2Isct32Msg |
| iK4Y-JKRRAc |
| lSCEt_mCHlM |
| ly1upu2FNTs |
| nxL0yqWP3H0 |
| pLJ7bC5Vcqw |
| rXFlJbXyZyc |
| tjqCzVjojCo |

test set:

| video_id |
| -------- |
| XIx-C22Ewk4 |
| bnW1PXGt5hw |

#### Changed label

Some original video_id shows the problem in the csv file, where we changed label name, it is not affect to use data set. If you want get original YouTube identifier, remove the prefix 'issue'.

train set:

| original video_id | changed video_id |
| ----------------- | ---------------- |
| -FaXLcSFjUI | issue-FaXLcSFjUI |
| -IELREHX_js | issue-IELREHX_js |

AVA-Dataset

AVA-Dataset 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

AVA-Dataset

AVA-Dataset 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README