Low latency object detection on the Edge-cloud AprilTag-assisted object detection and positioning
This study proposes a low-latency video processing pipeline for object detection and positioning. The pipeline employs GPU-based inferenceframeworks and lightweight models for fast detection. Moreover, twonovel low-error pose estimation algorithms are introduced, Multi-tagsaveraging (MTA) and Multi-...
Main Author: | |
---|---|
Format: | Others |
Language: | English |
Published: |
Uppsala universitet, Institutionen för informationsteknologi
2021
|
Subjects: | |
Online Access: | http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-446379 |
id |
ndltd-UPSALLA1-oai-DiVA.org-uu-446379 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-UPSALLA1-oai-DiVA.org-uu-4463792021-06-19T05:28:15ZLow latency object detection on the Edge-cloud AprilTag-assisted object detection and positioningengWang, DongUppsala universitet, Institutionen för informationsteknologi2021Engineering and TechnologyTeknik och teknologierThis study proposes a low-latency video processing pipeline for object detection and positioning. The pipeline employs GPU-based inferenceframeworks and lightweight models for fast detection. Moreover, twonovel low-error pose estimation algorithms are introduced, Multi-tagsaveraging (MTA) and Multi-points embedding (MPE), which reduce estimation error to 2 cm for 4K videos. You Only Calibrate Once (YOCO)is introduced for speeding up image recovering for distorted images. The whole pipeline is flexible and can be updated with faster objectdetection models or human pose estimation models in the future. The proposed pipeline achieves a latency of 41 ms while processing 4K videos on the task of object detection and positioning. Student thesisinfo:eu-repo/semantics/bachelorThesistexthttp://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-446379IT ; 21050application/pdfinfo:eu-repo/semantics/openAccess |
collection |
NDLTD |
language |
English |
format |
Others
|
sources |
NDLTD |
topic |
Engineering and Technology Teknik och teknologier |
spellingShingle |
Engineering and Technology Teknik och teknologier Wang, Dong Low latency object detection on the Edge-cloud AprilTag-assisted object detection and positioning |
description |
This study proposes a low-latency video processing pipeline for object detection and positioning. The pipeline employs GPU-based inferenceframeworks and lightweight models for fast detection. Moreover, twonovel low-error pose estimation algorithms are introduced, Multi-tagsaveraging (MTA) and Multi-points embedding (MPE), which reduce estimation error to 2 cm for 4K videos. You Only Calibrate Once (YOCO)is introduced for speeding up image recovering for distorted images. The whole pipeline is flexible and can be updated with faster objectdetection models or human pose estimation models in the future. The proposed pipeline achieves a latency of 41 ms while processing 4K videos on the task of object detection and positioning. |
author |
Wang, Dong |
author_facet |
Wang, Dong |
author_sort |
Wang, Dong |
title |
Low latency object detection on the Edge-cloud AprilTag-assisted object detection and positioning |
title_short |
Low latency object detection on the Edge-cloud AprilTag-assisted object detection and positioning |
title_full |
Low latency object detection on the Edge-cloud AprilTag-assisted object detection and positioning |
title_fullStr |
Low latency object detection on the Edge-cloud AprilTag-assisted object detection and positioning |
title_full_unstemmed |
Low latency object detection on the Edge-cloud AprilTag-assisted object detection and positioning |
title_sort |
low latency object detection on the edge-cloud apriltag-assisted object detection and positioning |
publisher |
Uppsala universitet, Institutionen för informationsteknologi |
publishDate |
2021 |
url |
http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-446379 |
work_keys_str_mv |
AT wangdong lowlatencyobjectdetectionontheedgecloudapriltagassistedobjectdetectionandpositioning |
_version_ |
1719411537966268416 |