Adobe Sensei: Mastering Content and Data
https://blog.adobe.com/media_9ff7e0cee1348f3a7f9023f1c54ddac14952acd7.gif
This is a big day for Adobe. Not only are we hosting the largest MAX creativity conference ever – we are introducing Adobe Sensei, a framework and set of intelligent services built into the Adobe Experience Platform which dramatically improve the design and delivery of digital experiences. “Sensei,” “master” or “teacher” in Japanese, is the perfect term to describe the deep expertise in artificial intelligence (AI), machine learning and deep learning we have built into our platform.
AI and machine learning have become ubiquitous in the technology industry, and a lot of great work has been done to build out horizontal frameworks to solve large-scale problems – from accurate speech recognition to computer vision. Unlike these broad efforts, Adobe Sensei is unique in its approach to blending the science of data with the art of human creativity. It is focused solely and deeply on solving complex experience challenges in the markets we serve. It powers Creative Cloud, Adobe Document Cloud and Adobe Marketing Cloud, spanning our entire platform, to help our customers work smarter and faster.
Adobe Sensei brings together two unique Adobe capabilities combined with the latest technology advancements in AI, machine learning, deep learning and related fields:
- A massive volume of content and data, from high-resolution images to customer clicks. This anonymized information powers our AI/machine learning engine, and as each of us knows from our own consumer experiences — whether it’s Amazon recommendations or Google Waze traffic information — more data drives a better outcome. Our pure focus on digital experiences means we can deliver much better contextual results for our customers.
- Deep understanding of how our customers work. We’re leveraging decades of intellectual property built into Adobe’s flagship products, already used by millions of customers. We have best-in-class AI/machine learning expertise which is being applied to areas our customers care about such as image understanding and categorization, font recognition, marketing attribution and intelligent recommendations to name a few.
There are dozens of Adobe Sensei intelligent services already running within our cloud solutions, and with today’s announcement, we’re committing to significantly increase our investment in both the shared framework across the Adobe Experience Platform as well as an even greater volume of intelligent services moving forward.
Here is a depiction of how Adobe Sensei fits into the overall Adobe Experience Platform – a framework running across our content and data assets, with intelligent services for each cloud offering running on top.
http://blogs.adobe.com/conversations/files/2016/11/architecture_wbg.png
Adobe Sensei intelligent services across our cloud offerings include these, among many others:
Creative Cloud
Content intelligence: Uses deep learning to search and tag images automatically, and makes intelligent recommendations when a user searches for images
Font Recognition: Recognizes a font from a creative asset and recommends similar fonts, even from handwriting
Face Aware Editing: Finds faces in an image and uses “landmarks” such as eyebrows, lips and eyes to understand their position and change the facial expression without ruining the image
Semantic Segmentation: Shows each image region labeled with its type; for example, whether it is a building or the sky. Such labelled regions allow easy selection and manipulation of objects, using simple commands (e.g., “change the sky”)
New Creative Cloud features rolling out today at MAX built with Adobe Sensei include:
- Adobe Stock Visual Search: Allows you to “find images like this” by simply dragging any still image file (.jpeg or .png) to your browser.
- Face-Aware Liquify: Enables you to adjust a person’s facial features, including changing the look of their eyes, mouth, nose, chin and other facial features.
- Auto Lip Sync: Automatically converts spoken dialogue into cartoon mouth animation in real-time.
Adobe Marketing Cloud
Attribution: Algorithmically determines the impact of different marketing touch points on consumers’ decisions to engage with a brand, determines the effectiveness of different campaigns and makes optimized marketing investment recommendations
Personalization: Quickly identifies and recommends relevant offers, experiences and messages to truly engage visitors
Intelligent Audience Segmentation: Gives marketers and analysts new visibility into which segments are most important to their businesses and allows them to target overlapping or adjacent segments, making it possible to acquire customers much more efficiently
Anomaly Detection: Simplifies data analysis by surfacing the most relevant insights and highlighting anomalies
Sentiment Analysis: Helps see and predict the content that customers will like, talk about and share most
We announced new Adobe Marketing Cloud features powered by Adobe Sensei today. They include:
- Intelligent Alerts: Gives an automated message when something material is happening – for example, when smartphone cart abandonments see an irregular spike.
- Automated Insights for Digital Advertising: Provides automated analysis of the performance of ads that usually takes hours to create manually. For example, if you should invest more in search versus social or promote a campaign on a specific date or time.
- New Lookalike Modeling Capabilities: Allows brands to further personalize and target prospects that share the same interests and traits – such as identifying frequent fliers as an audience segment for promotions.
- One-click Personalization: Determines the highest-performing content options, such as different layouts, wording and images for individuals. With one click, brands can automatically determine the right website or app experience to show each visitor to drive better engagement and conversion.
Adobe Document Cloud
Natural Language Processing: Provides text understanding, topic modeling and sentiment analysis of digital documents
Semantic Structure Analysis: Analyzes scanned or plain documents to identify features such as paragraphs, tables and headings from visual features
Document Similarity: Recognizes similar documents for search and highlights their differences
Document Summarization: Abbreviates a document by identifying a small number of representative sentences
Optical Character Recognition: Converts scanned documents to structure and text
Document Flow Extraction: Automatically extracts reading order from structured documents, enabling reflow and accessibility solutions
More Adobe Document Cloud features built with Adobe Sensei will be rolling out in upcoming months.
Opening Adobe Sensei to Our Ecosystem
We plan to make the Adobe Sensei framework and intelligent services available to our ecosystem of partners, ISVs and developers to extend its capabilities and enable custom solutions. If you’re a developer, be sure to check out adobe.io where more information will be coming about how to build on top of Adobe Sensei.
This is Just the Beginning
In the future, imagine Adobe Sensei being able to dynamically arrange your Creative Cloud user experience, surfacing the tools others use to accomplish a similar creative task; or Adobe Document Cloud telling you which of your white papers is being shared the most; or Adobe Marketing Cloud alerting you to target a new audience for a marketing campaign that you didn’t even ask it to find. The possibilities are endless, and our teams are hard at work thinking about how to turn new ideas into services our customers can use every day.
We have always pushed the boundaries of what is possible in creating and delivering digital experiences, and with the rollout of Adobe Sensei we are recommitting to staying on the forefront. We’re excited to see where our customers and partners take it.
Follow @AdobeSensei and stay tuned to all Adobe Sensei updates.