Webinar Q&A: Automatically Analyzing Video with Watson and OpenWhisk

Share this post:

While video becomes more important as a digital media type, video data often remains dark to analytics. This webinar demonstrates how IBM Bluemix OpenWhisk together with Watson services begins to unlock the value of video data.

Dark Vision is an application that uploads video files or streams to the cloud, transcodes video data, extracts and passes frames through the Watson Image Recognition and the Alchemy Face Recognition services, and generates meta-data to use in categorizing the video data for searchability.

In the presentation, Andreas Nauerz introduces the basics of IBM Bluemix OpenWhisk and Frederic Lavigne demonstrates how to create OpenWhisk code to accomplish Dark Vision’s application workflow.

Featured Speakers

Dr. Andreas Nauerz – Technical Product Manager at the IBM Germany Research & Development Lab
Frederic Lavigne – Bluemix Developer Advocate

Questions & Answers

  • How does it work with Docker? It takes a second or two to start. Does that mean my request processing time will have this overhead?Yes, starting in Docker’s container can take a few seconds. The Whisk system will try to reuse containers, so most of the time on the first invocation it will take a few seconds. Subsequent invocations, if they happen pretty quickly, will not have to restart the container. It’ll reuse the extent container, so those will go pretty quick. It’s basically container caching.
  • What is the pricing for OpenWhisk? We kept saying again and again that you only paid what you used, but how does that actually work?Since we currently have started the experimental phase, where we do not charge here at all. We do not have finalized pricing models, so we do not have numbers than we can share with you. We understand that this is of interest, so we are currently evaluating and working on our pricing model. The only thing I can really tell that we will charge you really for what you use, which probably means that we will go for something like that we charge you for the number of actions and locations, and probably in correlation with something like the duration per action and the memory allocation per action. This is not like a commitment. This is more sharing what our results are. You need to wait a little bit until we have finalized our pricing model.
  • I want to enhance the Dark Vision app to transcribe the audio to text and associate video frames with recognized images and faces as a way of making video data searchable. How easily could that be done?We have the Watson speech to text module that could leverage in this app. That will definitely be a good enhancement to the app that could be done. Given the current structure of the code and how it’s built, that would be something fairly easy to add to the demo. Which, as I was commenting in the walkthrough before, we’ll be publishing that in GitHub. Still have some more to do around the code to clean it up before publishing it, but that will come. You will be able to do those enhancement and contribute them back.
  • We talked about image processing. We talked about text to speech, but I think this question is very interesting because they’re asking how can we leverage both text to speech and image at the same time? I don’t believe it does that today. Is that correct?No, it does not, It’s only about images. That presents a really interesting possibility for the future.
  • How does OpenWhisk fit into the Agile DevOps program?I think it’s the perfect technology, especially for companies shifting towards HR organization where they have a lot of small teams. Because the variety was a model that is very flexible. Each single team that has been associated with a particular task can reuse the existing skills because we support you with all these different languages you can stick to. You can reuse what you have learned in the past. Also allows you to tackle the problem that has been assigned to you using the best technology that is available there. If there is a problem that you can better tackle with using Swift, you can Swift.Maybe you want to do that because a particular team already has developed skills in this area, maybe because they do mobile development or have done mobile development in the past, and now want to outsource computer-intensive tasks. We really tried to allow each single team to stick what they’ve learned and what they already like.I think that also applies to, not only to the development process, but of course, also to the entire operational process. Especially when referring to the operations, one point I want to make clear again is that one of the key benefits here is that we hide any infrastructure operational complexity. We really make it easy for you to focus on developing value-adding code. We want to relieve you from worrying about its operational aspects. The entire aspect of operational efforts that need to be spent actually that should get less important when you go for OpenWhisk, because we want to relieve you from worrying about it.
  • What is the minimum requirement in terms of gigabytes, RAM for an on-premise installation of OpenWhisk? Is there a certain minimum number of VMs required? Now, they’re talking not about an OpenWhisk instance that is hosted by IBM, but one that’s hosted locally. For example, if they had a dedicated server.I remember that at the beginning, we had a requirement like 6 or 8 GB of RAM. There’s an open source version of OpenWhisk you can get from GitHub, and it runs comfortably in a 4 GB VM. That’s what we do our center testing on. OpenWhisk itself is implemented as a bunch of Docker containers and deployed in 4 GB VMs.
  • When will it be possible to fully utilize Swift on OpenWhisk? Most of the implementations we’ve seen would not be possible in Swift right now, as only a subset of languages support it, such as synchronous callbacks out of an action are not possible.Our Whisk implementation runs Swift on Linux, and so we’re tracking the latest drops of Swift on Linux from The question is correct that Swift on Linux does not have all the functionality that Swift on iOS has yet. Basically, we’re gated by the progress of the Swift language itself. We expect Swift 3 to be announced later this year, and we expect at that point Swift will be much more mature on Linux. In the meantime I would say we will track the latest drops of Swift on Linux. As far as understanding how that compares to iOS, I’ll refer you to for the Swift open source project
  • Is there a visual editor render application flows or services flows?If you sign up for the experimental program and you are approved, we won’t have a code editor or a visual editor out there at the moment. The only graphical interface that we have there is the dashboard, which shows you your activity log and the activations you had in the past. Of course, we are working on a code editor that will then provide functionality in addition to CLI, the command line interface that we already have. We expect that to be released very soon. I cannot tell you an exact release date, but we’ll definitely work on that. It’s pretty obvious that people are asking for that, and that they want that, and need that. We are working on that, and I expect that to be released very, very soon.
  • Is it possible to have an on-premises instance of Watson?I’m not aware of any. I believe it’s a little more demanding on resources than one that would be hosted locally.
More Events stories
May 3, 2019

Kubernetes Tutorials: 5 Ways to Get You Building Fast

Ready to start working with Kubernetes? Want to build your Kubernetes skills? The five tutorials in this post will teach you everything you need to know about how to manage your containerized apps with Kubernetes.

Continue reading

May 3, 2019

Using Portworx to Deploy and Manage an HA MySQL Cluster on IBM Cloud Kubernetes Service

This tutorial is a walkthrough of the steps involved in deploying and managing a highly available MySQL cluster on IBM Cloud Kubernetes Service.

Continue reading

May 2, 2019

Kubernetes v1.14.1 Now Available in IBM Cloud Kubernetes Service

We are excited to announce the availability of Kubernetes v1.14.1 for your clusters that are running in IBM Cloud Kubernetes Service. IBM Cloud Kubernetes Service continues to be the first public managed Kubernetes service to support the latest upstream versions from the community.

Continue reading