Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving
An innovative solution named Annotation as a Service (AaaS) has been specifically designed to integrate heterogeneous video annotation workflows into containers and take advantage of a cloud native highly scalable and reliable design based on Kubernetes workloads. Using the AaaS as a foundation, the...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-06-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/12/4301 |
| _version_ | 1797564379494875136 |
|---|---|
| author | Sergio Sánchez-Carballido Orti Senderos Marcos Nieto Oihana Otaegui |
| author_facet | Sergio Sánchez-Carballido Orti Senderos Marcos Nieto Oihana Otaegui |
| author_sort | Sergio Sánchez-Carballido |
| collection | DOAJ |
| description | An innovative solution named Annotation as a Service (AaaS) has been specifically designed to integrate heterogeneous video annotation workflows into containers and take advantage of a cloud native highly scalable and reliable design based on Kubernetes workloads. Using the AaaS as a foundation, the execution of automatic video annotation workflows is addressed in the broader context of a semi-automatic video annotation business logic for ground truth generation for Autonomous Driving (AD) and Advanced Driver Assistance Systems (ADAS). The document presents design decisions, innovative developments, and tests conducted to provide scalability to this cloud-native ecosystem for semi-automatic annotation. The solution has proven to be efficient and resilient on an AD/ADAS scale, specifically in an experiment with 25 TB of input data to annotate, 4000 concurrent annotation jobs, and 32 worker nodes forming a high performance computing cluster with a total of 512 cores, and 2048 GB of RAM. Automatic pre-annotations with the proposed strategy reduce the time of human participation in the annotation up to 80% maximum and 60% on average. |
| first_indexed | 2024-03-10T18:56:30Z |
| format | Article |
| id | doaj.art-964ffeb0632a4d1ab54d9a7b4f2bc378 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T18:56:30Z |
| publishDate | 2020-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-964ffeb0632a4d1ab54d9a7b4f2bc3782023-11-20T04:42:16ZengMDPI AGApplied Sciences2076-34172020-06-011012430110.3390/app10124301Semi-Automatic Cloud-Native Video Annotation for Autonomous DrivingSergio Sánchez-Carballido0Orti Senderos1Marcos Nieto2Oihana Otaegui3Department of Intelligent Transport Systems and Engineering, Vicomtech, Paseo Mikeletegi 57, 20009 Donostia/San Sebastián, SpainDepartment of Intelligent Transport Systems and Engineering, Vicomtech, Paseo Mikeletegi 57, 20009 Donostia/San Sebastián, SpainDepartment of Intelligent Transport Systems and Engineering, Vicomtech, Paseo Mikeletegi 57, 20009 Donostia/San Sebastián, SpainDepartment of Intelligent Transport Systems and Engineering, Vicomtech, Paseo Mikeletegi 57, 20009 Donostia/San Sebastián, SpainAn innovative solution named Annotation as a Service (AaaS) has been specifically designed to integrate heterogeneous video annotation workflows into containers and take advantage of a cloud native highly scalable and reliable design based on Kubernetes workloads. Using the AaaS as a foundation, the execution of automatic video annotation workflows is addressed in the broader context of a semi-automatic video annotation business logic for ground truth generation for Autonomous Driving (AD) and Advanced Driver Assistance Systems (ADAS). The document presents design decisions, innovative developments, and tests conducted to provide scalability to this cloud-native ecosystem for semi-automatic annotation. The solution has proven to be efficient and resilient on an AD/ADAS scale, specifically in an experiment with 25 TB of input data to annotate, 4000 concurrent annotation jobs, and 32 worker nodes forming a high performance computing cluster with a total of 512 cores, and 2048 GB of RAM. Automatic pre-annotations with the proposed strategy reduce the time of human participation in the annotation up to 80% maximum and 60% on average.https://www.mdpi.com/2076-3417/10/12/4301computing scalabilityvideo annotationkubernetesADASautonomous drivingground truth data annotation |
| spellingShingle | Sergio Sánchez-Carballido Orti Senderos Marcos Nieto Oihana Otaegui Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving Applied Sciences computing scalability video annotation kubernetes ADAS autonomous driving ground truth data annotation |
| title | Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving |
| title_full | Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving |
| title_fullStr | Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving |
| title_full_unstemmed | Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving |
| title_short | Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving |
| title_sort | semi automatic cloud native video annotation for autonomous driving |
| topic | computing scalability video annotation kubernetes ADAS autonomous driving ground truth data annotation |
| url | https://www.mdpi.com/2076-3417/10/12/4301 |
| work_keys_str_mv | AT sergiosanchezcarballido semiautomaticcloudnativevideoannotationforautonomousdriving AT ortisenderos semiautomaticcloudnativevideoannotationforautonomousdriving AT marcosnieto semiautomaticcloudnativevideoannotationforautonomousdriving AT oihanaotaegui semiautomaticcloudnativevideoannotationforautonomousdriving |