EfficientSign: An Attention-Enhanced Lightweight Architecture for Indian Sign Language Recognition

2026-04-09

AI Summary

This study addresses the challenge of Indian Sign Language alphabet recognition on resource-constrained mobile devices by proposing a lightweight deep learning model. Built upon an EfficientNet-B0 backbone, the architecture integrates both channel-wise (Squeeze-and-Excitation) and spatial attention mechanisms to enhance focus on discriminative gesture features. The method operates without manual feature engineering and achieves a remarkable accuracy of 99.94% on a dataset of 12,637 images with only 4.2 million parameters—reducing parameter count by 62% compared to ResNet18. It substantially outperforms conventional approaches such as SURF and existing lightweight models, demonstrating both high efficiency and strong potential for real-world deployment on edge devices.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Learning on the Edge & Model Compression
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

How do you build a sign language recognizer that works on a phone? That question drove this work. We built EfficientSign, a lightweight model which takes EfficientNet-B0 and focuses on two attention modules (Squeeze-and-Excitation for channel focus, and a spatial attention layer that focuses on the hand gestures). We tested it against five other approaches on 12,637 images of Indian Sign Language alphabets, all 26 classes, using 5-fold cross-validation. EfficientSign achieves the accuracy of 99.94% (+/-0.05%), which matches the performance of ResNet18's 99.97% accuracy, but with 62% fewer parameters (4.2M vs 11.2M). We also experimented with feeding deep features (1,280-dimensional vectors pulled from EfficientNet-B0's pooling layer) into classical classifiers. SVM achieved the accuracy of 99.63%, Logistic Regression achieved the accuracy of 99.03% and KNN achieved accuracy of 96.33%. All of these blow past the 92% that SURF-based methods managed on a similar dataset back in 2015. Our results show that attention-enhanced learning model provides an efficient and deployable solution for ISL recognition without requiring a massive model or hand-tuned feature pipelines anymore.

Problem

• Indian Sign Language Recognition
• lightweight model
• mobile deployment
• hand gesture recognition
• efficient architecture

Innovation

• lightweight architecture
• attention mechanism
• Indian Sign Language recognition
• EfficientNet
• model efficiency