Segment object in new domain using SAM and making input features domain-agnostic with APSeg
Segment object in new domain using SAM and making input features domain-agnostic with APSeg
APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic Segmentation
arXiv paper abstract https://arxiv.org/abs/2406.08372
arXiv PDF paper https://arxiv.org/pdf/2406.08372
Few-shot semantic segmentation (FSS) endeavors to segment unseen classes with only a few labeled samples … performances degrade … applied to a distinct domain.
… propose to leverage … Segment Anything Model (SAM) … introduce APSeg, a novel auto-prompt network for cross-domain few-shot semantic segmentation (CD-FSS), which is designed to be auto-prompted for guiding cross-domain segmentation.
… propose a Dual Prototype Anchor Transformation (DPAT) module that fuses pseudo query prototypes extracted based on cycle-consistency with support prototypes, allowing features to be transformed into a more stable domain-agnostic space.
Additionally, a Meta Prompt Generator (MPG) module is introduced to automatically generate prompt embeddings, eliminating the need for manual visual prompts.
… build an efficient model which can be applied directly to target domains without fine-tuning.
… model outperforms the state-of-the-art CD-FSS method by 5.24% and 3.10% in average accuracy on 1-shot and 5-shot settings, respectively.
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website