Salvato in:
Dettagli Bibliografici
Autori principali: Mbobda-Kuate, Kwame, Kasmi, Gabriel
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2603.02142
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910038145105920
author Mbobda-Kuate, Kwame
Kasmi, Gabriel
author_facet Mbobda-Kuate, Kwame
Kasmi, Gabriel
contents Scaling laws assume larger models trained on more data consistently outperform smaller ones -- an assumption that drives model selection in computer vision but remains untested in resource-constrained Earth observation (EO). We conduct a systematic efficiency analysis across three scaling dimensions: model size, dataset size, and input resolution, on rooftop PV detection in Madagascar. Optimizing for model efficiency (mAP$_{50}$ per unit of model size), we find a consistent efficiency inversion: YOLO11N achieves both the highest efficiency ($24\times$ higher than YOLO11X) and the highest absolute mAP$_{50}$ (0.617). Resolution is the dominant resource allocation lever ($+$120% efficiency gain), while additional data yields negligible returns at low resolution. These findings are robust to the deployment objective: small high-resolution configurations are Pareto-dominant across all 44 setups in the joint accuracy-throughput space, leaving no tradeoff to resolve. In data-scarce EO, bigger is not just unnecessary: it can be worse.
format Preprint
id arxiv_https___arxiv_org_abs_2603_02142
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Is Bigger Always Better? Efficiency Analysis in Resource-Constrained Small Object Detection
Mbobda-Kuate, Kwame
Kasmi, Gabriel
Computer Vision and Pattern Recognition
Machine Learning
Scaling laws assume larger models trained on more data consistently outperform smaller ones -- an assumption that drives model selection in computer vision but remains untested in resource-constrained Earth observation (EO). We conduct a systematic efficiency analysis across three scaling dimensions: model size, dataset size, and input resolution, on rooftop PV detection in Madagascar. Optimizing for model efficiency (mAP$_{50}$ per unit of model size), we find a consistent efficiency inversion: YOLO11N achieves both the highest efficiency ($24\times$ higher than YOLO11X) and the highest absolute mAP$_{50}$ (0.617). Resolution is the dominant resource allocation lever ($+$120% efficiency gain), while additional data yields negligible returns at low resolution. These findings are robust to the deployment objective: small high-resolution configurations are Pareto-dominant across all 44 setups in the joint accuracy-throughput space, leaving no tradeoff to resolve. In data-scarce EO, bigger is not just unnecessary: it can be worse.
title Is Bigger Always Better? Efficiency Analysis in Resource-Constrained Small Object Detection
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2603.02142