Early-Exit Criteria for Edge Semantic Segmentation

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.

Authors

Institutions

Publication Details

Published
2025-05-26
DOI
https://doi.org/10.1109/icmlcn64995.2025.11140234
Primary Topic
Advanced Neural Network Applications
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Early-Exit Criteria for Edge Semantic Segmentation

Marcello L. R. de Campos, Marcelo Dias de Amorim, Anne Fladenmuller, Rodrigo S. Couto et al.
Advanced Neural Network Applications
article

Early-Exit Criteria for Edge Semantic Segmentation

Marcello L. R. de Campos, Marcelo Dias de Amorim, Anne Fladenmuller, Rodrigo S. Couto, Mateus S. Gilbert, Roberto G. Pacheco, Miguel Elias M. Campista
article en

Abstract

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.

Universidade Federal do Rio de Janeiro (BR), Centre National de la Recherche Scientifique (FR), Universidade Federal Fluminense (BR), Sorbonne Université (FR)
Bpifrance, Fundação de Amparo à Pesquisa do Estado de São Paulo, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Conselho Nacional de Desenvolvimento Científico e Tecnológico, Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro
Openalex Percentile: Top 77%
Advanced Neural Network Applications
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Early-Exit Criteria for Edge Semantic Segmentation — Marcello L. R. de Campos, Marcelo Dias de Amorim, et al. · (2025) | TGRS Research Map | TGRS