Calculating energy levels for kids' activities

Energy-Meter tool for dance and similar exercises

The Client

Advanced Companion Robotics Startup

One of the most innovative and advanced robotics startups, developing an AI-based robot to educate and entertain children wanted us to add another feature of Energy Meter for dance and other exercises in their robots.

Akaike’s edge-cloud agnostic solutions offer great flexibility.

Flexibility 0%
Cost 0%

Executive Summary

Industry Overview


The Consumer Robotics Market size was valued at USD 5.17 Billion in 2020 and is projected to reach USD 41.78 Billion by 2028, growing at a CAGR of 29.83% from 2021 to 2028. The Global Consumer Robotics Market has grown substantially in recent times as robotic automation is gaining popularity. Consumer task robots imitate human actions to do tasks in the same way humans perform. The application of robotics to consumer tasks enables achieving time, cost, and comfort benefits. Furthermore, the technology provides consumers with added advantages, such as reduced effort and better peace of mind, which is expected to drive the demand in the Global Consumer Robotics Market during the forecasted period.
Business Challenge

Customer Experience

This successful AI-based companion robot startup was looking to integrate a feature of Energy Meter for Dance and other similar movement-driven exercises in their devices. The energy meter would help the user understand the effort the child puts in while dancing or exercising. It requires the device’s camera to capture all the explosive movements done during the dance to assign an energy score. The energy score was to be calculated based on the effort the user puts into the dance or exercise; i.e., the higher the intensity and the effort a user puts in, the higher the energy score. It would also be used to help kids participate in a healthy competition with fellow kids dancing together in front of the device. Hence, it was required to detect each child separately and provide individual energy scores.

The Akaike Edge

Inbuilt libraries, DL models with transfer learning capabilities

Experienced ML and DL Ops teams

Efficient Deployment 0%
Integration 0%
Ongoing Maintenance 0%


Inbuilt libraries, DL models with transfer learning capabilities
Step 1.

Defining an energy score for the energy meter

Energy estimation is a function of the rate of change of a person's position and gestures from one frame to another. Videos are typically captured at 24/30 FPS and the approach required only a subset of the frames sampled at equal time intervals to compute the energy. The reduction in the frame consumption for energy computation helped in reducing the requirement for heavier hardware.

Step 2.

Selecting the appropriate pose estimator and fetching the landmark points

Our solution used a lightweight pose estimation model to capture the poses for each individual in the frame and compute the energy as a factor of the effort. Various iterations with Media-pipe, mmPose, movenet, Posenet, and NCNN-based pose-estimators were done to improve performance on the video frames, and the estimated landmark points were further fed to the energy-computing function.

Step 3.

Deploying an edge solution with an android application

The deployment was on an edge device which required building an android application that takes video frames as input and renders energy on the screen. Despite the hardware being lightweight and only capable of executing single-pose estimators, we were able to simulate multi-pose estimators for the simultaneous calculation of energies for two users.

Research shows that 87% of Vision AI
projects do not yield expected results

either owing to training data insufficiencies stalling the project, or being too slow to deploy. Our AI experts can help you accelerate in data sparse environments.

We use cookies to give you the best experience.

Get in Touch