Hugging Face's logo

Datasets:
avery00
/
MomentSeeker

Tasks:
Question Answering
Languages:
English
ArXiv:
License:
Dataset card Data Studio Files Community
2
MomentSeeker
File size: 8,689 Bytes 
068762b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
10c2c1b
 
 
068762b
 
 
10c2c1b
 
068762b
 
2b4babe
068762b
10dddf4
 
 
068762b
4aa0263
2b4babe
068762b
 
 
 
 
 
 
069b5dc
068762b
 
 
 
 
 
 
 
bfa8e36
 
068762b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
62cc3dc
068762b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2b4babe
068762b
 
 
 
 
 
 
 
 
 
 
 
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
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
 
---
license: cc-by-nc-sa-4.0
extra_gated_prompt: >-
  You acknowledge and understand that: This dataset is provided solely for
  academic research purposes. It is not intended for commercial use or any other
  non-research activities. All copyrights, trademarks, and other intellectual
  property rights related to the videos in the dataset remain the exclusive
  property of their respective owners. 
   You assume full responsibility for any additional use or dissemination of this dataset and for any consequences that may arise from such actions. You are also aware that the copyright holders of the original videos retain the right to request the removal of their videos from the dataset. 
   Furthermore, it is your responsibility to respect these conditions and to use the dataset ethically and in compliance with all applicable laws and regulations. Any violation of these terms may result in the immediate termination of your access to the dataset.
extra_gated_fields:
  Institution: text
  Name: text
  Country: country
  I want to use this dataset for:
    type: select
    options:
    - Research
  I agree to use this dataset solely for research purposes: checkbox
  I will not use this dataset in any way that infringes upon the rights of the copyright holders of the original videos, and strictly prohibit its use for any commercial purposes: checkbox
task_categories:
- question-answering
language:
- en
---
<h1 align="center">MomentSeeker: A Comprehensive Benchmark and A Strong Baseline For Moment Retrieval Within Long Videos</h1>
<p align="center">
  <a href="https://yhy-2000.github.io/MomentSeeker/">
      <img alt="Homepage" src="https://img.shields.io/badge/🌐 Homepage-MomentSeeker-blue">
  </a>
    <a href="https://arxiv.org/abs/2502.12558">
            <img alt="Build" src="http://img.shields.io/badge/cs.CV-arXiv%3A2406.04264-B31B1B.svg">
    </a>
    <a href="https://huggingface.co/datasets/avery00/MomentSeeker">
        <img alt="Build" src="https://img.shields.io/badge/🤗 Dataset-MomentSeeker Benchmark-yellow">
    </a>
</p>

This repo contains the annotation data for the paper "[MomentSeeker: A Comprehensive Benchmark and A Strong Baseline For Moment Retrieval Within Long Videos](https://arxiv.org/abs/2502.12558)".



## 🔔 News:
- 🥳 2025/03/07: We have released the MomentSeeker [Benchmark](https://huggingface.co/datasets/avery00/MomentSeeker) and [Paper](https://arxiv.org/abs/2502.12558)! 🔥

## License
Our dataset is under the CC-BY-NC-SA-4.0 license.

⚠️ If you need to access and use our dataset, you must understand and agree: **This dataset is for research purposes only and cannot be used for any commercial or other purposes. The user assumes all effects arising from any other use and dissemination.**

We do not own the copyright of any raw video files. Currently, we provide video access to researchers under the condition of acknowledging the above license. For the video data used, we respect and acknowledge any copyrights of the video authors. Therefore, for the movies, TV series, documentaries, and cartoons used in the dataset, we have reduced the resolution, clipped the length, adjusted dimensions, etc. of the original videos to minimize the impact on the rights of the original works. 

If the original authors of the related works still believe that the videos should be removed, please contact [email protected] or directly raise an issue.


## Introduction

We present MomentSeeker, a comprehensive benchmark to evaluate retrieval models' performance in handling general long-video moment retrieval (LVMR) tasks. MomentSeeker offers three key advantages. First, it incorporates long videos of over 500 seconds on average, making it the first benchmark specialized for long-video moment retrieval. Second, it covers a wide range of task categories (including Moment Search, Caption Alignment, Image-conditioned Moment Search, and Video-conditioned Moment Search) and diverse application scenarios (e.g., sports, movies, cartoons, and ego), making it a comprehensive tool for assessing retrieval models' general LVMR performance. Additionally, the evaluation tasks are carefully curated through human annotation, ensuring the reliability of assessment. We further fine-tune an MLLM-based LVMR retriever on synthetic data, which demonstrates strong performance on our benchmark. The checkpoint will release soon.



![Illustrative overview of our MomentSeeker dataset.](https://cdn-uploads.huggingface.co/production/uploads/66d916a7b86f0d569aa19b60/ff-9bFKlN466wElhiA4Wi.png)




## 🏆 Mini Leaderboard
| Rank | Method                                    | Backbone         | # Params | CA      | MS      | IMS     | VMS     | Overall |
|------|------------------------------------------|-----------------|---------|--------|--------|--------|--------|--------|
| 1    | **V-Embedder**                            | InternVideo2-Chat| 8B       | <u>42.2</u> | **20.4** | **15.0** | **15.8** | **23.3** |
| 2    | CoVR                                    | BLIP-Large       | 588M     | 25.8    | 17.4    | <u>12.3</u> | <u>12.3</u> | <u>17.1</u> |
| 3    | InternVideo2                            | ViT              | 1B       | **44.6** | <u>18.2</u> | 4.8     | 0.0     | 16.9    |
| 4    | MM-Ret                                  | CLIP-Base        | 149M     | 23.2    | 15.4    | 10.5    | 10.5    | 14.9    |
| 5    | LanguageBind                            | CLIP-Large       | 428M     | 39.6    | 16.4    | 3.2     | 0.0     | 14.8    |
| 6    | E5V                                     | LLaVA-1.6        | 8.4B     | 25.8    | 16.8    | 6.2     | 5.2     | 13.5    |
| 7    | UniIR                                   | CLIP-Large       | 428M     | 25.0    | 15.2    | 6.4     | 0.0     | 10.9    |
| 8    | MLLM2VEC                                | LLaVA-1.6        | 8.4B     | 6.4     | 6.2     | 3.0     | 3.0     | 4.7     |
| 9    | MagicLens                               | CLIP-Large       | 428M     | 9.0     | 2.4     | 3.2     | 2.8     | 4.4     |



## License
Our dataset is under the CC-BY-NC-SA-4.0 license.

⚠️ If you need to access and use our dataset, you must understand and agree: **This dataset is for research purposes only and cannot be used for any commercial or other purposes. The user assumes all effects arising from any other use and dissemination.**

We do not own the copyright of any raw video files. Currently, we provide video access to researchers under the condition of acknowledging the above license. For the video data used, we respect and acknowledge any copyrights of the video authors. Therefore, for the movies, TV series, documentaries, and cartoons used in the dataset, we have reduced the resolution, clipped the length, adjusted dimensions, etc. of the original videos to minimize the impact on the rights of the original works. 

If the original authors of the related works still believe that the videos should be removed, please contact [email protected] or directly raise an issue.


## Evaluation
> Before you access our dataset, we kindly ask you to thoroughly read and understand the license outlined above. If you cannot agree to these terms, we request that you refrain from downloading our video data.

The JSON file provides candidate videos for each question. The candidates can be ranked, and metrics such as Recall@1 and MAP@5 can be computed accordingly.



## Hosting and Maintenance
The annotation files will be permanently retained. 

If some videos are requested to be removed, we will replace them with a set of video frames sparsely sampled from the video and adjusted in resolution. Since **all the questions in MomentSeeker are only related to visual content** and do not involve audio, this will not significantly affect the validity of MomentSeeker (most existing MLLMs also understand videos by frame extraction).

If even retaining the frame set is not allowed, we will still keep the relevant annotation files, and replace them with the meta-information of the video, or actively seek more reliable and risk-free video sources.





## Citation

If you find this repository useful, please consider giving a star 🌟 and citation

```
@misc{yuan2025momentseekercomprehensivebenchmarkstrong,
      title={MomentSeeker: A Comprehensive Benchmark and A Strong Baseline For Moment Retrieval Within Long Videos}, 
      author={Huaying Yuan and Jian Ni and Yueze Wang and Junjie Zhou and Zhengyang Liang and Zheng Liu and Zhao Cao and Zhicheng Dou and Ji-Rong Wen},
      year={2025},
      eprint={2502.12558},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2502.12558}, 
}
```