Newest papers

Multiclass Boundary Refinement with Neural Cellular Automata for Remote Sensing Land Cover Segmentation

Pascoassis Souza Santos Meira, Gilson Antonio Giraldi,  Rafael Duarte Coelho dos Santos · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.30007 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

A segmentação semântica de imagens de detecção remota muitas vezes coloca limites de classe a alguns pixels de distância de onde eles realmente estão, e este erro de posição é concentrado na faixa fina medida pela IoU de fronteira (BIoU). Tratamos o refinamento de fronteira como uma aplicação iterada de uma pequena regra local aprendida: o Limite Refinamento Neural Cellular Automaton (BRNCA), que começa com as probabilidades de classe de um segmentador congelado, pré-treinado e as corrige célula por célula. Generalizamos o rNCA binário para o caso multiclasse mantendo os canais visíveis no simples de probabilidade através de uma atualização Aitchison (log-space) e adicionando canais ocultos livres. A melhor configuração melhora a BIo média

Machine translation (Meta M2M100); the English original follows.

Abstract

Semantic segmentation of remote sensing imagery often places class boundaries a few pixels away from where they truly lie, and this position error is concentrated in the thin band measured by Boundary IoU (BIoU). We treat boundary refinement as the iterated application of a small learned local rule: the Boundary Refinement Neural Cellular Automaton (BRNCA), that starts from the class probabilities of a frozen, pre-trained segmenter and corrects them cell by cell. We generalise the binary rNCA to the multiclass case by keeping the visible channels on the probability simplex through an Aitchison (log-space) update and by adding free hidden channels. The best configuration improves the mean BIoU over the base segmenter by 2.36 points (from 46.63 to 48.99). The gain is concentrated on the irregular woodland class (+8.5 points BIoU); this suggests that position error is correlated with the fractal dimension of the class boundaries.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Boundary (topology), Computer science, Segmentation, Artificial intelligence, Remote sensing, Pixel

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange