HIGHLIGHTS
SUMMARY
The same a uthors30, first used a medium-sized end-to-end neural_network model, namely a dual time-scale convolutional neural_network (DTSCNN), to recognize micro-expressions. Since the sample size of micro-expression datasets is still small, they cannot be adequately trained using deep convolutional neural_networks, and the training cost is high, making it difficult for ordinary | Vol:(1234567890) 12:17522 | Micro‑expression recognition model design Following is a description of the proposed model. The microexpression dataset did not contain sufficient data, and the number of expressions in different categories was extremely unbalanced after . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.