Towards Edge Computing Using Early-Exit Convolutional Neural Networks

Autor(es)

Pacheco, Roberto G. ^[1] ; Bochie, Kaylani ^[1] ; Gilbert, Mateus S. ^[1] ; Couto, Rodrigo S. ^[1] ; Campista, Miguel Elias M. ^[1]

Afiliação do(s) autor(es)

^[1] Univ Fed Rio de Janeiro UFRJ, Grp Teleinformat & Automacao GTA, PEE COPPE DEL Poli, BR-21941972 Rio De Janeiro - Brazil

• Tipo de documento: Artigo Científico
• Fonte: INFORMATION; v. 12, n. 10 OCT 2021.
• Citações Web of Science: 0

Resumo

In computer vision applications, mobile devices can transfer the inference of Convolutional Neural Networks (CNNs) to the cloud due to their computational restrictions. Nevertheless, besides introducing more network load concerning the cloud, this approach can make unfeasible applications that require low latency. A possible solution is to use CNNs with early exits at the network edge. These CNNs can pre-classify part of the samples in the intermediate layers based on a confidence criterion. Hence, the device sends to the cloud only samples that have not been satisfactorily classified. This work evaluates the performance of these CNNs at the computational edge, considering an object detection application. For this, we employ a MobiletNetV2 with early exits. The experiments show that the early classification can reduce the data load and the inference time without imposing losses to the application performance.</p> (AU)

• Processo FAPESP: 15/24494-8 - Comunicação e processamento de big data em nuvens e névoas computacionais
• Beneficiário: Nelson Luis Saldanha da Fonseca
• Modalidade de apoio: Auxílio à Pesquisa - Temático