K-centered Patch Sampling for Efficient Video Recognition

For decades, it has been a common practice to choose a subset of video frames for reducing the computational burden of a video understanding model. In this paper, we argue that this popular heuristic might be sub-optimal under recent transformer-based models. Specifically, inspired by that transform...

Full description

Saved in:
Bibliographic Details
Published inComputer Vision - ECCV 2022 Vol. 13695; pp. 160 - 176
Main Authors Park, Seong Hyeon, Tack, Jihoon, Heo, Byeongho, Ha, Jung-Woo, Shin, Jinwoo
Format Book Chapter
LanguageEnglish
Published Switzerland Springer 2022
Springer Nature Switzerland
SeriesLecture Notes in Computer Science
Subjects
Online AccessGet full text

Cover

Loading…