Image Segmentation Comparison Hub
3 papers - avg viability 6.7
Top Papers
- StructSAM: Structure- and Spectrum-Preserving Token Merging for Segment Anything Models(7.0)
StructSAM accelerates Segment Anything Model (SAM) inference by intelligently merging tokens, offering a speedup with minimal accuracy loss, making it ideal for real-time segmentation applications.
- Spatio-Semantic Expert Routing Architecture with Mixture-of-Experts for Referring Image Segmentation(7.0)
SERA enhances referring image segmentation by introducing a novel expert routing architecture for improved spatial coherence and boundary precision.
- Visualizing Coalition Formation: From Hedonic Games to Image Segmentation(6.0)
Utilize coalition formation in hedonic games for image segmentation, offering a novel approach to pixel grouping and boundary detection.