There’s no doubt about it-Deep Learning-based image analysis has come to the fore. Deep learning is a branch of artificial Intelligence that identifies patterns in data. That data could be images, sounds, the list is endless. What caused this rise? Simply put, the increased and enhanced performance of computers and associated algorithms.
Take a look around you-multiple organizations are investing heavily on building artificial intelligence capabilities to improve upon algorithms for image analysis capabilities. In fact, it isn’t a secret that several startups are also expanding in the artificial Intelligence domain and are working on building image analysis competencies.
A few of the organizations have started offering image analysis capabilities in the form of readily available services over open APIs. These organizations are continuously refining the algorithms and upgrading their image catalogue so as to increase the performance and accuracy of the results and avoid any overhead on application developers.
The convenience of the image analysis capabilities over open interfaces have brought immense opportunities for many organizations to come up with diverse features to their existing content-based products and applications. This, naturally, provides an innovative edge to products and enhances the end users experience. The integration of these capabilities through open interfaces into the existing products and applications reduces the development life cycle, speeds the time to market and at the same time enriches the product experiences through innovative features.
Just to give a glimpse of how the image analysis can provide an edge in existing products or help in building innovative product altogether, some exciting product enhancements are listed over here. I am sure the list may go endless but they should be adequate to give you a prevue of how the applications can be enriched by using the AI capabilities.
- Applications supporting Image based search
One of an attention-grabbing use case is to add the image based search to the current set of content based applications. If you see a picture and just desires to do search based on that, no necessity to enter that object description or celebrity name etc. You just need to open the application interface to click at the image search option, point the camera to the object/artist or the famous person that you want to get information for and just click at it.
- Adding image moderation capability to content management systems
In the content management system, the content providers can upload the content, associated meta-data and then share the content/metadata to different applications for consumption. Through image moderation capability, it is possible to notice the ‘explicit’ content which may not be appropriate for some applications.
- Adding more authenticity to the attendance management system
One way to improve the attendance systems are to let the attendance system pre store the images of employees and compare the image of the person punching the card to the pre-stored data and check if the card punched indeed corresponds to the person punching it.
- Image detection based surveillance
The surveillance system can detect the movement, takes images and compares it to the fed information in the system. If it detects some movement that does not compare to the fed images, it raises alarms.
- Kids’ Education Applications
In the coming years, image analysis can play a huge factor in education sector and help children learn quickly by using their phone, pointing it to different objects and get to know about them various details.
Suggestion from my end for the organizations exploring to get the image analysis based functionality into their products, is to gauge the market by quickly integrating the functionality by creating prototypes, demo it to the intended customers and getting their feedback to fine tune and eventually make their products future ready.
One thing is for sure that the image analysis based experiences are there to stay. Different organization will come up with different innovative ideas to build their product portfolio more robust. Need to wait for some time before it makes inroads into our life like many other innovative things that we are now get used to of.
‘’Image analysis capabilities are increasingly becoming advantageous for providing cutting edge landscapes to the applications in the content based product portfolio. The features like discovering specific content in the Ring Back Tone application through artist/singer image, adding moderation capabilities to content management systems, filtering user generated content for any ‘explicit’ content and enhancing Augmented Reality based application are some of the upcoming innovations in the Artificial Intelligence domain, noted Manish Jain, Head of Engineering, Digital Lifestyle Solutions at Mahindra Comviva.’’
Manish Jain, Head of Engineering, Digital Lifestyle Solutions at Mahindra Comviva.
Manish brings to the table over 17 years of work experience in the telecom value added services Industry, spanning research and development projects and product conceptualization and development. A core techie at heart, he has filed for over 20 patents that include inventions around network nodes and handset-based solutions and finance. He holds a bachelor’s degree in Electronics and Communication from Delhi College of Engineering, as well as an MBA in Operations.