<Z/>
ZahidulDEV
AI & Robotics ResearchResearch Published
2024

Social Interactive Robot

Human-Robot Interaction & Real-Time Computer Vision Platform

48 FPS

Frame Rate

22ms

Inference Latency

94.2%

Detection Accuracy

120+

Tested Scenarios

PROJECT OVERVIEW

What was built?

A cutting-edge Human-Robot Interaction (HRI) project integrating spatial-temporal deep learning models to enable autonomous social robots to perceive, analyze, and react to human gestures and facial expressions in real time.

THE CHALLENGE

Problem Statement

Conventional robotic vision systems often struggle with lighting variance, occlusion, and real-time inference latency. In social HRI environments, delayed response times disrupt natural human communication and degrade user trust.

THE SOLUTION

Engineering Approach & Architecture

We developed a lightweight MobileNetV3 + Bi-LSTM neural network architecture connected via WebSocket streams to a responsive web dashboard. This achieved 48 FPS real-time processing while running on low-power edge GPUs.

System Architecture Data Flow
01Camera Feed Capture -> OpenCV Frame Preprocessing
02Feature Extraction -> Lightweight MobileNetV3 Convolutional Backbone
03Temporal Modeling -> Bidirectional LSTM Sequence Encoder
04Decision Pipeline -> Action Classification & Robotic Arm Command Signal
05Telemetry Dashboard -> Real-time WebSocket Data Visualization
PROBLEM SOLVING

Key Engineering Obstacles

Obstacle: High Latency During Multi-Person Tracking

Implemented frame-skipping heuristics and dynamic region-of-interest (ROI) cropping to prioritize the primary speaker, reducing GPU memory bandwidth by 40%.

Obstacle: False Positives in Low-Light Conditions

Applied adaptive histogram equalization and temporal smoothing across 5-frame rolling windows to stabilize detection confidence scores.

Technologies & Tools Used

Core libraries, frameworks, and deployment services.

PythonOpenCVPyTorchROS / AIReactWebSocket
Social Interactive Robot Case Study — Zahidul Islam | Zahidul Islam