AI RESEARCH
ClustViT: Clustering-based Token Merging for Semantic Segmentation
arXiv CS.CV
•
ArXi:2510.01948v2 Announce Type: replace Vision Transformers can achieve high accuracy and strong generalization across various contexts, but their practical applicability on real-world robotic systems is limited due to their quadratic attention complexity. Recent works have focused on dynamically merging tokens according to the image complexity. Token merging works well for classification but is less suited to dense prediction. We propose ClustViT, where we expand upon the Vision Transformer (ViT) backbone and address semantic segmentation.