PANC: Prior-Aware Normalized Cut for Training-Free In-Context Segmentation
cs.CV, cs.AI
Submitted: 2026-02-06
Updated: 2026-11-18
Code: https://github.com/jgnav/PANC
Terminology
Sources
- Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture
- Emerging Properties in Self-Supervised Vision Transformers
- Upsampling DINOv2 features for unsupervised vision tasks and weakly supervised materials segmentation
- Masked Autoencoders Are Scalable Vision Learners
- Finding an Unsupervised Image Segmenter in Each of Your Deep Generative Models
- Localizing Objects with Self-Supervised Transformers and no Labels
- DINOv3
- TokenCut: Segmenting Objects in Images and Videos with Self-supervised Transformer and Normalized Cut
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs