Fusion of Multimodal Embeddings for Ad-Hoc Video Search
Abstract
The challenge of Ad-Hoc Video Search (AVS) originates from free-form (i.e., no pre-defined vocabulary) and free-style (i.e., natural language) query description. Bridging the semantic gap between AVS queries and videos becomes highly difficult as evidenced from the low retrieval accuracy of AVS benchmarking in TRECVID. In this paper, we study a new method to fuse multimodal embeddings which have been derived based on completely disjoint datasets. This method is tested on two datasets for two distinct tasks: on MSR-VTT for unique video retrieval and on V3C1 for multiple videos retrieval.
Cite
Text
Francis et al. "Fusion of Multimodal Embeddings for Ad-Hoc Video Search." IEEE/CVF International Conference on Computer Vision Workshops, 2019. doi:10.1109/ICCVW.2019.00233Markdown
[Francis et al. "Fusion of Multimodal Embeddings for Ad-Hoc Video Search." IEEE/CVF International Conference on Computer Vision Workshops, 2019.](https://mlanthology.org/iccvw/2019/francis2019iccvw-fusion/) doi:10.1109/ICCVW.2019.00233BibTeX
@inproceedings{francis2019iccvw-fusion,
title = {{Fusion of Multimodal Embeddings for Ad-Hoc Video Search}},
author = {Francis, Danny and Nguyen, Phuong Anh and Huet, Benoit and Ngo, Chong-Wah},
booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
year = {2019},
pages = {1868-1872},
doi = {10.1109/ICCVW.2019.00233},
url = {https://mlanthology.org/iccvw/2019/francis2019iccvw-fusion/}
}