With 900 million images reviewed daily, Tuputech technology learns and can judge images quickly, processing 50 images per second and identifying pornographic or violent images with 99.5% accuracy
Guangzhou, China – October 5, 2017 – To manage brands and social platforms in today’s online communities, companies must have a way to quickly and easily monitor shared content. Today Tuputech is introducing its cloud-based image recognition solution to the U.S. to help companies identify, tag and filter objectionable images—pornographic, violent or otherwise explicit or inappropriate. Using its state-of-the-art algorithms and computer vision, and processing 50 images per second, Tuputech provides real-time image recognition for social media, live-broadcasting and video and photo-sharing sites.
Based on deep learning and AI technology, Tuputech teaches and trains its software to learn and judge in a way similar to the way a human’s convolutional neural network (CNN) operates. With more clients and more images, the better the Tupu platform becomes. Tupu clients upload images and videos to the cloud where the software can quickly detect different types of inappropriate content such as nudity and violence.
Today Tuputech’s image recognition API can quickly and accurately detect objectionable images and videos so companies can reduce online auditing manpower, in some cases by more than 90%. Since the company started processing images three years ago when it was founded in China, Tupu’s average daily use of its API has surpassed 1 billion images, providing it with the huge database that is a prerequisite for generating accurate recognition results. It now handles 900 million images a day, and with over 100 billion images reviewed to date, Tuputech’s solution has a 95% accuracy rate.
“Interacting with a brand or social site online should be enjoyable. We shouldn’t have to worry about offensive content popping up, “ said Leonard Mingqiang, founder of Tuputech (and a former founding member of WeChat). “Our technology acts fast to detect inappropriate content and help our clients remove it before it enters the online environment.”
Tuputech’s image moderation and recognition solution flags the inappropriate content for review and handling by the client. It is already working with top companies in China such as Musical.ly, Live.me, China Mobile Communications, Thunder, China Unicom, and Viva Video to name a few. At Musical.ly the company has eliminated 95% of its human labor efforts to manage uploaded images and videos and today only 1% of their videos now require manual review.
“Our goal is to remove some of the stress and pressure that comes with the day-to-day operations of running a massive online business,” continued Mingqiang. “We take online image and content review off your plate. A company’s manual review becomes 1% of their time so they can focus on more important things and keep moving their company forward.”
Tuputech uses deep learning and AI and image recognition to flag objectionable content online. With a database of more than 1 billion images, the cloud-based technology learns to identify pornographic or violet images. Founded in 2014, the company is based in Guangzhou, China and has raised $10 million Series A funding from Northern Light Venture Capital and Morningside Venture Capital. For more information, visit us.tuputech.com.