LATCH: Learned arrangements of three patch codes

Gil Levi, Tal Hassner

פרסום מחקרי: פרק בספר / בדוח / בכנספרסום בספר כנסביקורת עמיתים

תקציר

We present a novel means of describing local image appearances using binary strings. Binary descriptors have drawn increasing interest in recent years due to their speed and low memory footprint. A known shortcoming of these representations is their inferior performance compared to larger, histogram based descriptors such as the SIFT. Our goal is to close this performance gap while maintaining the benefits attributed to binary representations. To this end we propose the Learned Arrangements of Three Patch Codes descriptors, or LATCH. Our key observation is that existing binary descriptors are at an increased risk from noise and local appearance variations. This, as they compare the values of pixel pairs: changes to either of the pixels can easily lead to changes in descriptor values and compromise their performance. In order to provide more robustness, we instead propose a novel means of comparing pixel patches. This ostensibly small change, requires a substantial redesign of the descriptors themselves and how they are produced. Our resulting LATCH representation is rigorously compared to state-of-the-art binary descriptors and shown to provide far better performance for similar computation and space requirements.

שפה מקוריתאנגלית
כותר פרסום המארח2016 IEEE Winter Conference on Applications of Computer Vision, WACV 2016
מוציא לאורInstitute of Electrical and Electronics Engineers Inc.
מסת"ב (אלקטרוני)9781509006410
מזהי עצם דיגיטלי (DOIs)
סטטוס פרסוםפורסם - 23 מאי 2016
אירועIEEE Winter Conference on Applications of Computer Vision, WACV 2016 - Lake Placid, ארצות הברית
משך הזמן: 7 מרץ 201610 מרץ 2016

סדרות פרסומים

שם2016 IEEE Winter Conference on Applications of Computer Vision, WACV 2016

כנס

כנסIEEE Winter Conference on Applications of Computer Vision, WACV 2016
מדינה/אזורארצות הברית
עירLake Placid
תקופה7/03/1610/03/16

הערה ביבליוגרפית

Publisher Copyright:
© 2016 IEEE.

טביעת אצבע

להלן מוצגים תחומי המחקר של הפרסום 'LATCH: Learned arrangements of three patch codes'. יחד הם יוצרים טביעת אצבע ייחודית.

פורמט ציטוט ביבליוגרפי