Spatial-temporal FAST corner detector for videos

Chia Chun Lee, Yun Jung Chang, Yen-Yu Lin, Chun Rong Huang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


Due to the increase of large-scale videos, efficient and effective corner detectors are necessary to facilitate video processing and the underlying applications. Designing a good corner detector often faces an accuracy/efficiency trade-off problem. To address this problem, a highly efficient and effective approach named spatial-temporal FAST corner detector is proposed to detecting corners for videos. Our approach considers both the spatial and temporal evidence to extract corners, each of which is an extreme when comparing it with the pixels on the surface of the spatial-temporal tube centered on it. To efficiently obtain spatial-temporal FAST corners, a two-step procedure is adopted. The FAST detector [1] is firstly used to filter out most non-corner pixels. Then, temporal consistency is checked for FAST corners to identify if these corners are spatial-temporal FAST corners. As a result, the detected corners are spatial-temporal coherent. In the experiments, seven the state-of-the-art corner detectors are compared. The results show the superiority of our approach in both repeatability and efficiency.

Original languageEnglish
Title of host publicationIntelligent Systems and Applications - Proceedings of the International Computer Symposium, ICS 2014
EditorsWilliam Cheng-Chung Chu, Stephen Jenn-Hwa Yang, Han-Chieh Chao
PublisherIOS Press
Number of pages11
ISBN (Electronic)9781614994831
StatePublished - 1 Jan 2015
EventInternational Computer Symposium, ICS 2014 - Taichung, Taiwan
Duration: 12 Dec 201414 Dec 2014

Publication series

NameFrontiers in Artificial Intelligence and Applications
ISSN (Print)0922-6389


ConferenceInternational Computer Symposium, ICS 2014


  • corner detection
  • spatial-temporal information
  • video processing

Fingerprint Dive into the research topics of 'Spatial-temporal FAST corner detector for videos'. Together they form a unique fingerprint.

Cite this