Web-based Video-Assisted Point Cloud Annotation for ADAS validation

Mujika, Andoni; Fanlo, Ana Dominguez; Tamayo, Inigo; Senderos, Orti; Barandiarán, Javier Avilés; Aranjuelo, Nerea; Nieto, Marcos; Otaegui, Oihana

doi:10.1145/3329714.3338128

Cited by 7 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, even if there are different users working with the tool or defining several use cases, the terms are chosen from the ontology and remain the same for every use case. For instance, in the case of web-based annotation tools [25], its functionality could be boosted by including relations among the annotated objects. In addition, the previous annotations and configuration files may be checked according to the vocabulary defined in the ontology, enhancing interoperability by ensuring a common understanding of the domain concepts.…”

Section: Semantic Labelingmentioning

confidence: 99%

Design and Implementation of an Ontology for Semantic Labeling and Testing: Automotive Global Ontology (AGO)

et al. 2021

Self Cite

View full text Add to dashboard Cite

Modern Artificial Intelligence (AI) methods can produce a large quantity of accurate and richly described data, in domains such as surveillance or automation. As a result, the need to organize data at a large scale in a semantic structure has arisen for long-term data maintenance and consumption. Ontologies and graph databases have gained popularity as mechanisms to satisfy this need. Ontologies provide the means to formally structure descriptive and semantic relations of a domain. Graph databases allow efficient and well-adapted storage, manipulation, and consumption of these linked data resources. However, at present, there is no a universally defined strategy for building AI-oriented ontologies for the automotive sector. One of the key challenges is the lack of a global standardized vocabulary. Most private initiatives and large open datasets for Advanced Driver Assistance Systems (ADASs) and Autonomous Driving (AD) development include their own definitions of terms, with incompatible taxonomies and structures, thus resulting in a well-known lack of interoperability. This paper presents the Automotive Global Ontology (AGO) as a Knowledge Organization System (KOS) using a graph database (Neo4j). Two different use cases for the AGO domain ontology are presented to showcase its capabilities in terms of semantic labeling and scenario-based testing. The ontology and related material have been made public for their subsequent use by the industry and academic communities.

show abstract

Section: Semantic Labelingmentioning

confidence: 99%

Design and Implementation of an Ontology for Semantic Labeling and Testing: Automotive Global Ontology (AGO)

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Inverse perspective mapping IPM were used in [1], [2] to segment the road image into classes for region of interest, vehicles and background. First a prior handling stage was executed, in this the image quality was investigated and if the quality was low it was enhanced.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The inverse perspective mapping's main objective is to place and remap every-pixel concerning a diverse position, to produce an original two-dimensional array of pixels and to facilitate this task of the vehicle detection and the road modeling [2], [17], [18]. This technique provides a upper view image, as shown in the Figure 5.…”

Section: Inverse Perspective Mappingmentioning

confidence: 99%

See 1 more Smart Citation

Moving Object Tracking using Inverse Perspective Mapping and Optical Flow

Almutairi¹

2020

IJATCSE

View full text Add to dashboard Cite

As objects moving at a pace create difficulty in identification so is the problem persisting in the moving vehicle recognition and tracking using passive vision sensor. This proposed tracking system is based on a camera mounted in a moving vehicle used to capture images to track another moving one. It considers tracking vehicles in open roads and highways. Inverse perspective mapping is adopted to extract a bird eye view of the road to subtract the standpoint results from the image. The image is then cropped and processed to extract the vehicle segment. The cropped image is returned to the camera view where the optical flow is utilized to estimate the vehicle direction and relative velocity based on the object's centroid displacement. The algorithm is tested on standard datasets as well as a new dataset created in Egypt's open roads. The results show robust tracking even with curved roads and fast computation which allow real-time processing even on embedded systems with low-computing resources.

show abstract

“…Some tools allow one to annotate a polygon when it is convenient (e.g., 3D objects) [16], keypoints (e.g., human pose [17]), temporal events [18], or the association of objects between frames of the video [12,17]. Semi-automatic approaches exist to optimize the annotation process by reducing human interactions e.g, with recognition of human activity [19], detection of objects and segmentation throughout a video [20][21][22], as well as methods based on single frame detection such as [23][24][25]. Semi-automatic means they still need HiL interacting in some part of the annotation process.…”

Section: Introductionmentioning

confidence: 99%

Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving

et al. 2020

Self Cite

View full text Add to dashboard Cite

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.

show abstract

Web-based Video-Assisted Point Cloud Annotation for ADAS validation

Cited by 7 publications

References 8 publications

Design and Implementation of an Ontology for Semantic Labeling and Testing: Automotive Global Ontology (AGO)

Design and Implementation of an Ontology for Semantic Labeling and Testing: Automotive Global Ontology (AGO)

Moving Object Tracking using Inverse Perspective Mapping and Optical Flow

Semi-Automatic Cloud-Native Video Annotation for Autonomous Driving

Contact Info

Product

Resources

About