Skip to main content

Table 6 Effects of different strategies on the results

From: ANINet: a deep neural network for skull ancestry estimation

  Convolutional network Accuracy (%) Para size
Local depth images AlexNet 90.83  − 7.21 45 MB  − 4 k
SE-AlexNet (SE block) 92.83  − 5.21 45 MB
Wide-AlexNet(Wide) 96.58  − 1.46 45 MB -4 k
ANINet(SE-Wide) 98.04 45 MB
Global depth image AlexNet 95.04  − 3.17 45 MB  − 4 k
SE-AlexNet(SE block) 95.93  − 2.28 45 MB
Wide-AlexNet(Wide) 97.47  − 0.74 45 MB  − 4 k
ANINet(SE-Wide) 98.21 45 MB
Local + global depth image AlexNet 95.56  − 3.47 45 MB  − 4 k
SE-AlexNet(SE block) 97.91  − 1.12 45 MB
Wide-AlexNet (Wide) 98.43  − 0.60 45 MB  − 4 k
ANINet (SE-Wide) 99.03 45 MB
  1. The significance of bold means that in a certain aspect (performance, model size), this model is the best among these models