Artificial Intelligence is the answer to many unanswered questions. This emerging technology has been applied in numerous sectors, among which healthcare and retail have benefited the most. Use cases such as chatbots, price optimisation, demand forecasting, data analysis, fraud detection, among others have been used heavily in these two sectors.
Over the past couple of years, computer vision technology has come a long way in helping businesses innovate successfully. It makes the most of existing cameras as well as data and enables business-altering implementations. When we specifically think of the hospitality industry, one can’t help but wonder about the various ways in which autonomous solutions can be put in place for better customer experience and higher hygiene standards across the sector.
A New Delhi-based startup is leveraging AI with the intent of enabling hospitality and retail businesses to monitor operations by detecting and tracking anomalies in standard operating procedures (SOPs).
Analytics India Magazine caught up with Adit Chhabra, CEO and co-founder of Wobot.ai to gain insights on how this company is leveraging AI in retail and healthcare. Founded in 2017 by Tapan Dixit (COO), Tanay Dixit (CPO) and Adit Chhabra (CEO), Wobot.ai is an AI-powered computer vision SaaS platform which helps hospitality, food, retail, and manufacturing businesses monitor their operations.
According to Chhabra, the plug-and-play tool connects to any existing CCTV or other forms of cameras and helps detect and track anomalies in the SOPs. It currently offers various modules which include hygiene, food safety, pilferage, and customer experience. Last year, the startup collaborated with Indian Railway Catering and Tourism Corporation (IRCTC) to introduce AI to monitor the food production process across its several base kitchens. Wobot deploys solutions and pre-trained models for retail businesses and helps them capture key areas to generate customer insights.
With Wobot’s video analytics, it is also possible to detect and measure losses due to employee theft at the counter. Using our proprietary transaction detection algorithm, retailers can also get details of the actual transactions that took place.
The Wobot.ai platform includes architectures for human re-identification and activity tracking, object and posture recognition.
Object Detection: Object detection algorithms typically use extracted features and learning algorithms to recognize instances of an object category. Some use cases of object detection include face recognition, object tracking, and anomaly detection.
Posture Detection: Posture detection is done by determining three-dimensional orientations by tracking the movement and orientation of a body with respect to custom axes. Its use cases include real-time posture detection, Kinect and 3D camera based body posture detection.
Activity Recognition: Vision-based action recognition helps deduce human actions in the present state, and predict human actions in the future state based upon incomplete action executions.
Person REID: This feature uses Spatio-temporal and other body features to re-identify personnel.
Use of AI/ML @ Wobot.ai
Several machine learning techniques which include person REID, activity recognition coupled with semi or fully unsupervised learning has been utilised to build models. According to Chhabra, automating SOPs is one of Wobot’s core pillars and its proprietary activity recognition and person re-identification architecture coupled with unsupervised learning is revolutionising the video analytics sector. The machine learning algorithms developed at Wobot.ai can detect any deviations in SOPs and automatically list them and make them trackable for all relevant stakeholders in an organisation.
Size of Engineering Team
Currently, the size of the team is over 35 across the offices, and the team comprises design, engineering, data tagging, sales, and customer support people. A majority of the team comprises of engineers.
Tackling Talent Crunch
Chhabra said, “While hiring we look for people who are passionate about building what we are trying to do with Wobot. Transparency is key, and we are always open about our expectations from a particular role. It helps to be open when you are hiring potential candidates.” he added, “We also actively look within our networks; it always helps to have a common ground.”
Wobot.ai is currently working on adding more use cases within the hospitality sector as well as the manufacturing sector. Further, the company is looking forward to collaborating with relevant partners and hope to start scaling in North America and the GCC region as well.
Provide your comments below
A Technical Journalist who loves writing about Machine Learning and Artificial Intelligence. A lover of music, writing and learning something out of the box. Contact: firstname.lastname@example.org