Enhancing Embedded Object Tracking : A Hardware Acceleration Approach for Real-Time Predictability
While Siamese object tracking has witnessed significant advancements, its hard real-time behaviour on embedded devices remains inadequately addressed. In many application cases, an embedded implementation should not only have a minimal execution latency, but this latency should ideally also have zero variance, i.e., be predictable. This study aims to address this issue by meticulously analysing real-time predictability across different components of a deep-learning-based video object tracking system. Our detailed experiments not only indicate the superiority of Field-Programmable Gate Array (FPGA) implementations in terms of hard real-time behaviour but also unveil important time predictability bottlenecks. We introduce dedicated hardware accelerators for key processes, focusing on depth-wise cross-correlation and padding operations, utilizing high-level synthesis (HLS). Implemented on a KV260 board, our enhanced tracker exhibits not only a speed up, with a factor of 6.6, in mean execution time but also significant improvements in hard real-time predictability by yielding 11 times less latency variation as compared to our baseline. A subsequent analysis of power consumption reveals our approach's contribution to enhanced power efficiency. These advancements underscore the crucial role of hardware acceleration in realizing time-predictable object tracking on embedded systems, setting new standards for future hardware-software co-design endeavours in this domain.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2024 |
---|---|
Erschienen: |
2024 |
Enthalten in: |
Zur Gesamtaufnahme - volume:10 |
---|---|
Enthalten in: |
Journal of imaging - 10(2024), 3 vom: 13. März |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Zhang, Mingyang [VerfasserIn] |
---|
Links: |
---|
Themen: |
Deep learning |
---|
Anmerkungen: |
Date Revised 29.03.2024 published: Electronic Citation Status PubMed-not-MEDLINE |
---|
doi: |
10.3390/jimaging10030070 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM370246993 |
---|
LEADER | 01000caa a22002652 4500 | ||
---|---|---|---|
001 | NLM370246993 | ||
003 | DE-627 | ||
005 | 20240330001655.0 | ||
007 | cr uuu---uuuuu | ||
008 | 240328s2024 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.3390/jimaging10030070 |2 doi | |
028 | 5 | 2 | |a pubmed24n1355.xml |
035 | |a (DE-627)NLM370246993 | ||
035 | |a (NLM)38535150 | ||
035 | |a (PII)70 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Zhang, Mingyang |e verfasserin |4 aut | |
245 | 1 | 0 | |a Enhancing Embedded Object Tracking |b A Hardware Acceleration Approach for Real-Time Predictability |
264 | 1 | |c 2024 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Revised 29.03.2024 | ||
500 | |a published: Electronic | ||
500 | |a Citation Status PubMed-not-MEDLINE | ||
520 | |a While Siamese object tracking has witnessed significant advancements, its hard real-time behaviour on embedded devices remains inadequately addressed. In many application cases, an embedded implementation should not only have a minimal execution latency, but this latency should ideally also have zero variance, i.e., be predictable. This study aims to address this issue by meticulously analysing real-time predictability across different components of a deep-learning-based video object tracking system. Our detailed experiments not only indicate the superiority of Field-Programmable Gate Array (FPGA) implementations in terms of hard real-time behaviour but also unveil important time predictability bottlenecks. We introduce dedicated hardware accelerators for key processes, focusing on depth-wise cross-correlation and padding operations, utilizing high-level synthesis (HLS). Implemented on a KV260 board, our enhanced tracker exhibits not only a speed up, with a factor of 6.6, in mean execution time but also significant improvements in hard real-time predictability by yielding 11 times less latency variation as compared to our baseline. A subsequent analysis of power consumption reveals our approach's contribution to enhanced power efficiency. These advancements underscore the crucial role of hardware acceleration in realizing time-predictable object tracking on embedded systems, setting new standards for future hardware-software co-design endeavours in this domain | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a FPGA | |
650 | 4 | |a deep learning | |
650 | 4 | |a embedded system | |
650 | 4 | |a hardware acceleration | |
650 | 4 | |a high-level synthesis | |
650 | 4 | |a object tracking | |
650 | 4 | |a real-time system predictability | |
650 | 4 | |a siamese network | |
700 | 1 | |a Van Beeck, Kristof |e verfasserin |4 aut | |
700 | 1 | |a Goedemé, Toon |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Journal of imaging |d 2016 |g 10(2024), 3 vom: 13. März |w (DE-627)NLM26971538X |x 2313-433X |7 nnns |
773 | 1 | 8 | |g volume:10 |g year:2024 |g number:3 |g day:13 |g month:03 |
856 | 4 | 0 | |u http://dx.doi.org/10.3390/jimaging10030070 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 10 |j 2024 |e 3 |b 13 |c 03 |