Abstract: Deep learning models for medical image segmentation often struggle with task-specific characteristics, limiting their generalization to unseen tasks with new anatomies, labels, or modalities ...
Abstract: Three-dimensional (3D) medical image segmentation typically demands extensive labeled training samples, which is prohibitively time-consuming and requires significant expertise. Although ...
Token Painter: Training-Free Text-Guided Image Inpainting via Mask Autoregressive Models (AAAI 2026)
Token Painter is a training-free text-guided image inpainting model built upon the Mask AutoRegressive (MAR) paradigm. Unlike diffusion-based methods that model the entire image in latent space and ...
3 Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, China. 4 DFH Satellite Co., Ltd., Beijing, China. For PyTorch dependency, see ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results