Communication Dans Un Congrès Année : 2025

Early-Exit Criteria for Edge Semantic Segmentation

Critère de Sortie Anticipée pour la Segmentation Sémantique Edge

Résumé

Early-exit deep neural networks (EE-DNNs) are multi-output DNNs designed for resource-constrained and latency-sensitive implementations, leveraging auxiliary output layers to partition processing between local, edge, and cloud devices. In the case of semantic segmentation, the literature lacks an efficient exit policy to stop the inference process earlier. Our contributions to fill this gap are twofold: (i) an adaptation of the normalized entropy (NE)-based exit criterion and (ii) a region-based exit criterion that compares segmentation from consecutive early exits. Our analyses reveal that the region-based approach better suits semantic segmentation because it exploits the intermediate early exits more efficiently. Our experiments show that a region-based EE-DNN delivers the same mean intersection over union as an EE-DNN with NE, saving at least 630 million floating-point operations per image on average.

Fichier principal
Vignette du fichier
2025_ICMLCN_Gilbert-8.pdf (1.42 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04963398 , version 1 (24-02-2025)

Licence

Identifiants

Citer

Mateus S. Gilbert, Roberto G. Pacheco, Rodrigo S. Couto, Anne Fladenmuller, Marcelo Dias de Amorim, et al.. Early-Exit Criteria for Edge Semantic Segmentation. 2025 IEEE International Conference on Machine Learning for Communications and Networking (ICMLCN), May 2025, Barcelona, Spain. pp.1-6, ⟨10.1109/ICMLCN64995.2025.11140234⟩. ⟨hal-04963398⟩
269 Consultations
339 Téléchargements

Altmetric

Partager

  • More