摘要
We participated in one of the two types of Instance Search task in TRECVID 2019: Fully Automatic Search, without any human intervention. Firstly, the specific person and action are searched separately, and then we re-rank the two sorts of search results by ranking the one type scores according to the other type, as well as the score fusion. And thus, three kinds of final instance search results are submitted. Specifically, for the person search, our baseline consists of face detection, alignment and face feature selection. And for the action search, we integrate person detection, person tracking and feature selection into a framework to get the final 3D features for all tracklets in video shots. The official evaluations showed that our best search result gets the 4th place in the Automatic search.
| 源语言 | 英语 |
|---|---|
| 出版状态 | 已出版 - 1 1月 2019 |
| 已对外发布 | 是 |
| 活动 | 2019 TREC Video Retrieval Evaluation, TRECVID 2019 - Gaithersburg, 美国 期限: 12 11月 2019 → 13 11月 2019 |
会议
| 会议 | 2019 TREC Video Retrieval Evaluation, TRECVID 2019 |
|---|---|
| 国家/地区 | 美国 |
| 市 | Gaithersburg |
| 时期 | 12/11/19 → 13/11/19 |
学术指纹
探究 'Inf@TRECVID 2019: Instance search task' 的科研主题。它们共同构成独一无二的学术指纹。引用此
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver