INDEX

Artificial Intelligence/Machine Learning in Video Services

The video market is evolving on multiple fronts and facing challenges that range from maturation/stagnation to mobile first viewership. While much of these disparate fronts are regionally distributed there are core elements (e.g. rise in OTT content and services) that nearly universally applicable and companies operating throughout the video value chain and evolving to meet the needs of this changing landscape. In this report the market potential and impact of AI and machine learning (ML) within the video space is explored and analyzed.

The focus is on systems that have the capability to separate out elements of the video content, extract meaning, and interpret elements of the video. This work helps to add value to all forms of content, including enterprise video content, sports, news, major television and movie content, and short form content, by automating the metadata extraction, evaluating video based on specifically trained criteria, and automating processes such as highlight creation, which can occur based on intelligent understanding of specific types of scenes.

There are several key markets for AI/ML within the media markets:

  • Using AI to review video and generate meaningful metadata, and aligning that with business cases; this is the principal focus of this report.
  • The analytics opportunity centered around content and customer data and experiences
  • The video quality-related market, related to the encoding and consumer electronics markets

AI/ML is being used to automate and improve workflows within these market segments and in addition to operational efficiencies it can also reduce expenses (e.g. reduce need to staff up to accomplish tasks such as content curation/filtering or stat collecting). While the market is relatively nascent ABI Research anticipates the market will grow at a 58% CAGR from 2018 to 2022.  

Table of Contents

  • 1. ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING IN VIDEO
    • 1.1. Artificial Intelligence and Machine Learning Markets in Video
    • 1.2. Opportunities of Video AI
    • 1.3. Capturing Meaning From Videos
    • 1.4. Compute Complexity: Audio, Images, and Video
    • 1.5. Understanding AI and ML
  • 2. TECHNICAL AND BUSINESS ISSUES IN VIDEO AI/ML
    • 2.1. Compute Location and Data Locality
    • 2.2. Training Set Ownership
  • 3. MEDIA USE CASES
    • 3.1. Indexing
    • 3.2. Censorship, Compliance, and Ratings
    • 3.3. Enrichment/Improved Metadata
    • 3.4. Video Quality-Related AI
    • 3.5. Summarization, Previews, and Trailers
    • 3.6. Genre-Specific Use Cases
    • 3.7. Service-Type Use Cases
    • 3.8. Efficiency Improvements
  • 4. MEDIA EXAMPLES
    • 4.1. NFL Next Gen Stats
    • 4.2. U.S. Open Tennis
    • 4.3. ZoneTV
  • 5. VIDEO AI AS A SERVICE
    • 5.1. Dimensional Mechanics
    • 5.2. Valossa
    • 5.3. GrayMeta
    • 5.4. Google
    • 5.5. AWS
    • 5.6. Microsoft
    • 5.7. IBM Watson