File size: 1,977 Bytes
f190bb7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
924ac9c
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
# Activitynet-QA
The ActivityNet-QA dataset contains 58,000 human-annotated QA pairs on 5,800 videos derived from the popular ActivityNet dataset. The dataset provides a benckmark for testing the performance of VideoQA models on long-term spatio-temporal reasoning. 

## Splits
train: 32,000 QA pairs on 3,200 videos

val: 18,000 QA pairs on 1,800 videos

test: 8,000 QA pairs on 800 videos

## Question Format
All the questions are stored in the `*_q.json` files. Each entry in the json file is of the following format.

```
{
  "video_name": str, 
  "question": str, 
  "question_id": str
}
```
The `video_name` field corresponds to the orginal video id in the ActivityNet dataset, the url for this video is `https://www.youtube.com/watch?v=<video_name>`. The `question_id` field refer to the unique id for the question in the dataset. The `question` field contains the questions in English.


## Answer Format
All the answers are stored in the `*_a.json` files. Each entry in the file is of the following format. 

```
{
  "answer": str, 
  "type": int, 
  "question_id": str
}
```

The `answer` field contains the answer with respect to the question with `question_id`
The `type` file contains the question or answer types for this question: 

Question types: [0].Motion  [1].Spatial Relationship  [2].Temporal Relationship  [3-8].Free

Answer types: [3].Yes/No  [4].Color  [5].Object  [6].Location  [7].Number  [8].Other

## Licence

The code and the dataset are distributed under MIT LICENSE. They are only allowed for non-commercial use.

## Citation

If the project are helpful for your research, please cite

```
@inproceedings{yu2019activityqa,
    author = {Yu, Zhou and Xu, Dejing and Yu, Jun and Yu, Ting and Zhao, Zhou and Zhuang, Yueting and Tao, Dacheng},
    title = {ActivityNet-QA: A Dataset for Understanding Complex Web Videos via Question Answering},
    booktitle = {AAAI},
    pages = {9127--9134},
    year = {2019}
}
```


---
license: apache-2.0
---