TMCnet News

The 2015 Image Recognition Year in Review from GumGum Highlights Annual Technology Advancements in the Fast-Changing Field
[January 12, 2016]

The 2015 Image Recognition Year in Review from GumGum Highlights Annual Technology Advancements in the Fast-Changing Field


GumGum, the leading digital marketing platform for the visual web, today released The 2015 Image Recognition Year in Review report highlighting the myriad of groundbreaking technology achievements in the field of image recognition over the past twelve months. GumGum, whose solutions are powered by its advanced image recognition capabilities, chose to highlight the year's most intriguing developments in the field as it too has experienced enormous growth and change in 2015. This shared growth marks an increase in practical applications of image recognition across fields including marketing, security, and social media.

The report includes leading developments in image recognition this year such as:

  • Panasonic (News - Alert) revealed new smart home appliances that can see;
  • A research team at Carnegie-Mellon University demonstrated an iris scanner that can identify people from up to 40 feet away;
  • Wolfram developed a tool to identify any picture that is uploaded in just one word;
  • Researchers at MIT determined which Game of Thrones character is the most memorable using image recognition;
  • Facebook's (News - Alert) artificial intelligence teams developed an algorithm to recognize people even when their faces aren't visble;
  • Like a Shazam for plants, a researcher at Australia's Edith Cowan University determined the species of a plant using highly complex algorithms and artificial intelligence to scan the image for color, shape, and texture;
  • Dulight, a new portable device by Baidu, launched. This new product used image recognition and helped the visually-impaired see;
  • Pinterest launched a new way for users to search and purchase any product that is shown within an image or picture;
  • Google (News - Alert) announced its TensorFlow artificial intelligence engine allowing developers to innovate with its machine learning and image recognition tools;
  • And lastly, Microsoft won the sixth annual ImageNet Large Scale Visual Recognition Challenge. Microsoft's victory doubled the current record for correctly categorizing online images with only a 3.5 percent error rate.



"Image recognition is foundational to our business so we keep a sharp eye on what's happening in the field. The 2015 projects we've highlighted are fascinating examples of the field's rapid advancement within commercial and enterprise applications," said Ophir Tanz, CEO of GumGum. "At GumGum, we apply our image recognition capabilities to enable marketers and publishers to unlock untapped, or even unseen, value. We believe, however, that image recognition will have a transformational impact on the world and it's important that those advancements are recognized, including those coming out of academia."

For its part, GumGum's technology understands the content of images as well as their context. With its proprietary technology, GumGum initially pioneered in-image advertising delivering native units that monetize images for the world's leading publishers. With the addition of Mantii in 2015, GumGum now simultaneously unlocks new value within social images for marketers.


GumGum aims to maximize the visual web's potential, and marketers are benefiting from its efforts. Today, GumGum reaches over 400 million visitors as it views images and content across more than 2,000 premium publishers including Time Inc. (News - Alert), which GumGum announced an exclusive partnership with earlier this month.

About GumGum

As the inventor of in-image advertising, GumGum is the digital marketing platform for the visual web. Our technology unlocks the value of connected images, and delivers highly visible campaigns and rich insights to marketers. Reaching over 400 million visitors as they view images and content across more than 2,000 premium publishers, GumGum ads consistently achieve an 81% viewability rate and deliver 10 times better engagement than traditional display options, resulting in superior brand lift for marketers and increased revenue for publishers. Powered by its core image recognition technology, GumGum also offers Mantii, a real-time image recognition tool for brands to identify and activate their most important images and engage their top influencers on social media.

With GumGum, leading brands from AdAge's top 100 US advertising spenders list, including Disney (News - Alert), L'Oreal and Toyota, can target audiences with contextually relevant content through in-image and in-screen ads, native, video and programmatic options. GumGum is headquartered in Santa Monica, California, with six additional offices in the US and in the UK. Founded in 2007, GumGum is backed by investors including Morgan Stanley, NEA, Upfront Ventures, First Round Capital and Crosscut Ventures.


[ Back To TMCnet.com's Homepage ]