Stable Hyper-Pooling and Query Expansion for Event Detection

Matthijs Douze, Jerome Revaud, Cordelia Schmid, Herve Jegou

ICCV 2013

doi:10.1109/ICCV.2013.229 /iccv/2013/douze2013iccv-stable/

Abstract

This paper makes two complementary contributions to event retrieval in large collections of videos. First, we propose hyper-pooling strategies that encode the frame descriptors into a representation of the video sequence in a stable manner. Our best choices compare favorably with regular pooling techniques based on k-means quantization. Second, we introduce a technique to improve the ranking. It can be interpreted either as a query expansion method or as a similarity adaptation based on the local context of the query video descriptor. Experiments on public benchmarks show that our methods are complementary and improve event retrieval results, without sacrificing efficiency.

PDF ICCV Semantic Scholar

Cite

Text

Douze et al. "Stable Hyper-Pooling and Query Expansion for Event Detection." International Conference on Computer Vision, 2013. doi:10.1109/ICCV.2013.229

Markdown

[Douze et al. "Stable Hyper-Pooling and Query Expansion for Event Detection." International Conference on Computer Vision, 2013.](https://mlanthology.org/iccv/2013/douze2013iccv-stable/) doi:10.1109/ICCV.2013.229

BibTeX

@inproceedings{douze2013iccv-stable,
  title     = {{Stable Hyper-Pooling and Query Expansion for Event Detection}},
  author    = {Douze, Matthijs and Revaud, Jerome and Schmid, Cordelia and Jegou, Herve},
  booktitle = {International Conference on Computer Vision},
  year      = {2013},
  doi       = {10.1109/ICCV.2013.229},
  url       = {https://mlanthology.org/iccv/2013/douze2013iccv-stable/}
}