DiGSeg repurposes diffusion U-Nets as generalist segmentation learners by conditioning on image-mask latents and multi-scale CLIP text features, achieving strong cross-domain performance.
arXiv preprint arXiv:2408.12957 (2024)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
MAgSeg is a decoder-free MLLM approach that uses a new instruction-tuning format to segment complex smallholder agricultural landscapes directly from high-resolution satellite imagery.
citing papers explorer
-
Diffusion Model as a Generalist Segmentation Learner
DiGSeg repurposes diffusion U-Nets as generalist segmentation learners by conditioning on image-mask latents and multi-scale CLIP text features, achieving strong cross-domain performance.
-
MAgSeg: Segmentation of Agricultural Landscapes in High-Resolution Satellite Imagery using Multimodal Large Language Models
MAgSeg is a decoder-free MLLM approach that uses a new instruction-tuning format to segment complex smallholder agricultural landscapes directly from high-resolution satellite imagery.