Human-Computer Interaction with Hand Gesture Recognition Using ResNet and MobileNet

Table 6

Comparison of related work.

Reference	Feature extraction	Classification	Accuracy (%)	Number of classes
[28]	(i) CNN Model		90	31 class
[36]	(ii) CNN Model		88.87	28 class
[29]	(i) Hand shape-based description, where each hand image was described by a vector consisting of 15 values where the values represent the key points locations	(i) KNN classifier-Naïve-Bayesian	97.548	28 class
Our research study	(i) ResNet50 and MobileNetV2		98.2	32 class