ReqVidA - Requirements Video Analyzer

Introduction

ReqVidA - Logo

 

Download: Requirements Video Analyzer

In Software Engineering, requirements elicitation is an important factor for project success. Therefore, an intense communication with the customer right from the beginning of the project is necessary. On the basis of these meetings, the engineers will elaborate high-quality requirements. Therefore, the communication with the customer need to be documented very precisely.

Often, communicating with the customer happens in workshops which are structured and guided meetings with the customers influencing the project. Since the aim of workshops is a well-founded collaboration to elaborate the desired results, i.e. requirements, this is a well-established technique to get in contact with the stakeholders.

Mostly, a textual protocol brings together the results of the workshop. Noticing all important information is very difficult for the transcript writer. Therefore, video recording may help him to analyze the content of the meeting afterwards. Often, a workshop last some hours, wherefore the recorded meeting would be very long and has no specific connection to the written protocol. Thus, the received structure by both video-recording and protocol-writing is unintuitive and the relevant slots of the video are hard to find. The combined use of protocol and video may help to catch all information, but it does not simplify the post-processing of the workshop with respect to protocolling the contents.

At Leibniz University Hanover, a student developed a tool called Requirements Video Analyzer (ReqVidA) within his master thesis. The tool records a workshop on video and helps to create a protocol in the same time. Using annotations helps to mark relevant parts of the video and supports the accumulation with notes. This simplifies the structure and helps the reader to get an overview about the content of the meeting. The connection of video and notes reduces the difficulties with registering all the important information. Furthermore, it simplifies the further application of the video while writing the protocol.

Conception

ReqVidA supports the transcript writer within his tasks while writing the protocol. Therefore, we developed four main concepts (see figure 1) supporting both the record and writing during the meeting and the elaboration of the reporting after the workshop.

ReqVidA - Konzepte

Figure 1 - ReqVidA-Concepts

Below, the concepts are presented in detail:

1. Integration in workflow

To support the protocol writer in his tasks, functionalities like recording the video, adding annotations, and structuring the protocol, need to be included in the previous workflow without restricting the protocol writer. Therefore, we decided to use a single main view, which is adapted to the specific situation (execution or post-processing of the workshop). Thereby, the focus and the interactions of the protocol writer are not separated by different kind of views, which reduces the additional work while using the tool.

2. Creating structure

Another important aspect is to generate an intuitive structure by the system for both a single and several videos.

For various videos, we defined a structure for the project directory generated and managed by ReqVidA. Because of clear hierarchies and coherent identifier, navigating manually through the directories is possible.

We used the already existing concept of annotations to create the structure in a single video. A slot in the video identified to be relevant for future work may be marked with an annotation and with general information both within the workshop and afterwards. Thus, we get a visible connection between the notes and the video, which helps to find, identify, and elaborate a relevant spot. Therefore, it is not necessary to watch the whole video while post-processing the meeting. Using the annotations, the protocol writer can look at certain interesting spots. This reduces the effort and the barrier for re-using the video.

3. Flexible Annotation System

The system of annotations in ReqVidA defines the base-functionality to combine notes and videos. Because of the focus on the scenarios of utility, the system of annotations was developed for the use in requirement workshops. Therefore, the system provides annotations specifically for identification, elicitation, and elaboration of the requirements.

Concerning the further application in different kind of scenarios, the system of annotations can be adapted for the appliance of the tool by a specific user. The shortcuts for creating annotations may be allocated arbitrarily. Furthermore, the content of an annotation generated by a shortcut can be pre-defined to save time while creating and filling in the content of specific annotations.

Additionally, the user itself can define own annotations with specific attributes. Therefore, ReqVidA can be prepared individually for both different kinds of workshops and many other intended uses.

4. Automated Result Extraction

Caused by the wide choice of post-processing the annotations and the protocol, it is necessary to export one or more desired artefacts automatically. This option is essential to minimalize the later effort of the protocol writer by bringing together his elaborations.

The available structured dates enable the automatic export by the system. The user may specify his results using different pre-defined options. This offers the possibility to get lists or even the whole protocol with his prepared data.

Tool

The ReqVidA-System consists of two main components, Recorder and Analyzer.

Below, the components are presented.

Recorder

ReqVidA - Recorder

Figure 2 - ReqVidA Recorder

The Recorder is used to record and annotate the video image in realtime. During the video recording, he also offers the options to create and edit a protocol, a glossary and a stakeholder list.

Analyzer

ReqVidA - Analyzer

Figure 3 - ReqVidA Analyzer

The Analyzer is used for video analysis. He offers the same options as the Recorder for creating and editing annotations, a protocol, a glossary and a stakeholder list. Additionally, he has the option to extract snapshots out of the video.

Generally, it is provided that a recorded video with its associated data from the Recorder is loaded for further analysis into the Analyzer. However, the Analyzer is designed to evaluate any existing video, although the video was not recorded with the Recorder.

Contact

For technical questions, suggestions and problems are the following employees available.

Publications

  • 2015
  • Stephan Kiesling, Oliver Karras, Kurt Schneider: ReqVidA - Requirements Video Analyzer, In Gesellschaft für Informatik, Fachgruppentreffen Requirements Engineering, Windisch, 2015. Bibtex. Link

Further References

  • 2015
  • Pohl, Henning, Hettig, Markus, Karras, Oliver, Ötztürk, Hatice, Rohs, Michael: CapCouch: Home Control with a Posture-sensing Couch, In Adjunct Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2015 ACM International Symposium on Wearable Computers, pages 229-232, New York, NY, USA. ACM, 2015. Bibtex. Link