3D CNN METHOD AND APPARATUS FOR HIGH-SPEED IMAGE RECOGNITION USING 3D CONVOLUTIONAL NEURAL NETWORK

摘要

A high-speed image recognition method and apparatus using 3D CNN (3-dimension Convolutional Neural Network) are disclosed. A high-speed image recognition method using 3D CNN (3-dimension Convolutional Neural Network) includes inputting first image clips to 3D CNNs among image clips constituting an input image, and for each of the first image clips. Obtaining results obtained by calculating a softmax function through 3D CNN, calculating a score margin using the obtained results, and comparing the calculated score margin with a preset threshold And in response to the comparing step, determining whether to input the remaining image clips excluding the first image clips among the image clips constituting the input image into the 3D CNN. Therefore, it is possible to improve the calculation speed for image recognition.