Scalable action localization with kernel-space hashing - Archive ouverte HAL Access content directly
Conference Papers Year : 2015

Scalable action localization with kernel-space hashing


To detect and locate complex human actions in video, one trains a detector for each target class and applies it to the video content. This approach can scale to large video databases if the application of the detector can be made sublinear in the size of the database. Sublinear retrieval methods have been successfully explored for query-by-example but few were devised for these more challenging queries by detector. We put forward here a novel approximate search method that relies on LSH to support query-by-detector. We evaluate our method on a recent large action localization dataset and show it has significantly better efficiency than linear search.
No file

Dates and versions

hal-01436944 , version 1 (16-01-2017)



Andrei Stoian, Marin Ferecatu, Jenny Benois-Pineau, Michel Crucianu. Scalable action localization with kernel-space hashing. Image Processing (ICIP), 2015 IEEE International Conference on, Sep 2015, Québec, Canada. pp.257-261, ⟨10.1109/ICIP.2015.7350799⟩. ⟨hal-01436944⟩
50 View
0 Download



Gmail Facebook X LinkedIn More