Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{198568,
        author = {Imteyaz Shahzad and Amaan Sheikh and Waquas and Mhd. Hamza and Sheikh Kabir and Raymeen Pathan},
        title = {Discifit: On-Device Automated Exercise Repetition Counting for Gamified Mobile Fitness},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {9946-9953},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=198568},
        abstract = {Not many people move much these days. At the same time, eyes stay glued to screens more than ever. One answer comes in the form of an app called Discifit - it ties phone fun to real-world movement. Play only follows sweat. A built-in system tracks workouts using just the phone's camera. No extra gear needed. Nothing sent online. All processing happens right where it lands. Motion begins with basic body tracking tech - think BlazePose or MoveNet. That data flows into a BiLSTM model trained to spot timing patterns in motion. Joint angles get measured through geometry rules when possible. If visuals blur or lag hits, optical flow steps in quietly. Everything runs under one steady logic frame - like a silent referee watching reps. From earlier studies using camera and motion sensors for AERC, the idea for combining both approaches grow naturally. Built for Android and iOS, the method takes shape through practical steps laid out clearly here. Testing follows a set structure meant to keep results consistent. Speed versus size versus precision - each choice tips the balance differently when recognizing counts. Past work points strongly toward aiming higher than 95% correct counts as a solid goal.
Counting workout reps automatically finds motion patterns through body positioning tech. Instead of manual tracking, movement between frames helps detect each rep using pixel shifts. Body joint data feeds into a two-way memory network that learns timing in exercises. Smartphones handle calculations locally so user privacy stays intact during analysis. While screens often distract, they can also encourage activity when used thoughtfully. Games layered onto fitness routines shift attention from duration to engagement. Training models directly on gadgets avoid constant internet needs for feedback loops.},
        keywords = {automated exercise repetition counting, pose estimation, BiLSTM, optical flow, mobile health, screen-time management, gamification, on-device machine learning.},
        month = {April},
        }