"Hi Map", a mobile-end image labeling service provider

Absrtact: With the advent of the reading age, a large number of content generated in the pictures and video, the image recognition, understanding and analysis needs more and more intense. But before image recognition, depth learning, and artificial intelligence become truly reliable technologies, labeling seems like a

With the advent of the reading age, a large number of content generated in the picture and video, the image recognition, understanding and analysis of the need for more and more intense. But before the image recognition, the depth study, the artificial intelligence becomes the truly dependable technology, "marking" seems to be an inevitable transition. The former has to play the label started a nice, after a few users have just broken billion "beauty shoot", but also began to try "Picture + tag" social routines. For many of the apps that also have a lot of picture resources to settle but have never been mined effectively, it sounds like a good idea to have a "tagging" outsourcing service.

"Hi Picture" is a mobile end image labeling service provider, any app with a picture can be accessed through the SDK authorization mode, the app user can use the annotation service to release the picture with different annotation, the annotation form includes the text (interest, place, brand belong to the text content), voice, video, link and so on. For the cooperative app of Hi Graph, 1 days can complete the integration development, reduce the development cost, the later maintenance upgrade also by the Hi diagram.

Hi, founder Saturday told 36 Krypton, in the mobile internet age, pictures became the main media of people's expression and communication, which can carry more information than words can express. In the early stages of the picture explosion, picture filters, image landscaping, etc. are standing on the tuyere. In his view, the current accumulation of pictures has reached another level, has been to the algorithm and data mining stage, Hi diagram is the image of the core module to integrate open source, support all apps into the reading era. His fancy is the value of the picture in various fields, the current imagination mainly has the following several aspects:

① to create social closed-loop, so that the annotation information as a link between users, so as to link to the user more interested in people, things, objects, so as to increase the user activity of the app, improve the application of the overall app value.

In addition, and offline consumer scenes related to the label has to do O2O potential, if an app user bask in a certain brand, then this user is not only in the network community bask in a thing, still equivalent to the brand made a PR. This can play "share send a taxi ticket" mode: Users through the "Sun" brand, you can get the brand discount coupons, or get some brands around and so on. But, of course, after the app has some influence, it's going to take a look at how well the "picture tagging" is doing to improve the app's user viscosity.

② change the picture play, find more efficient and interesting way of reading, make it more simple, more stereoscopic. For example, the user through the picture + keyword + Voice + video + link simple combination to fully express their demands, while the reader in the reception of this information will be more lightweight.

For example, job seekers to do a resume, a picture may be a portal, the top link, video can jump to work. such as renting the app want to use a simpler way to describe the listing information, a picture plus a few keywords, a voice on it. Designers of the professional app, want to use the picture annotation in the corresponding place to give a clear description, while adding voice interpretation, and so on.

Hi figure on January 20 after the on-line, the intention to date more than 20 customers, customer types including electric shopping guide, vertical community, photo applications, IM, local O2O, smart home, 60% of customers in the vertical social class. Founder of Saturday University in Liaoning province, the first prize for college students Computer contest, co-founder Huang in Hi figure as technical director, has served in ZTE, co-founder Li Chengli as Marketing director, is the founder of the business logic website, but also the certification author of the Tiger sniffing net.