תקציר
Accuracy, descriptor size, and the time required for extraction and matching are all important factors when selecting local image descriptors. To optimize over all these requirements, this paper presents a CUDA port for the recent Learned Arrangement of Three Patches (LATCH) binary descriptors to the GPU platform. The design of LATCH makes it well suited for GPU processing. Owing to its small size and binary nature, the GPU can further be used to efficiently match LATCH features. Taken together, this leads to breakneck descriptor extraction and matching speeds. We evaluate the trade off between these speeds and the quality of results in a feature matching intensive application. To this end, we use our proposed CUDA LATCH (CLATCH) to recover structure from motion (SfM), comparing 3D reconstructions and speed using different representations. Our results show that CLATCH provides high quality 3D reconstructions at fractions of the time required by other representations, with little, if any, loss of reconstruction quality.
שפה מקורית | אנגלית |
---|---|
כותר פרסום המארח | Computer Vision – ECCV 2016 Workshops, Proceedings |
עורכים | Gang Hua, Herve Jegou |
מוציא לאור | Springer Verlag |
עמודים | 685-697 |
מספר עמודים | 13 |
מסת"ב (מודפס) | 9783319494081 |
מזהי עצם דיגיטלי (DOIs) | |
סטטוס פרסום | פורסם - 2016 |
אירוע | Computer Vision - ECCV 2016 Workshops, Proceedings - Amsterdam, הולנד משך הזמן: 8 אוק׳ 2016 → 16 אוק׳ 2016 |
סדרות פרסומים
שם | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
כרך | 9915 LNCS |
ISSN (מודפס) | 0302-9743 |
ISSN (אלקטרוני) | 1611-3349 |
כנס
כנס | Computer Vision - ECCV 2016 Workshops, Proceedings |
---|---|
מדינה/אזור | הולנד |
עיר | Amsterdam |
תקופה | 8/10/16 → 16/10/16 |
הערה ביבליוגרפית
Publisher Copyright:© Springer International Publishing Switzerland 2016.