Training-Free Semantic Segmentation via LLM-Supervision

arXiv:2404.00701v1 Announce Type: new
Abstract: Recent advancements in open vocabulary models, like CLIP, have notably advanced zero-shot classification and segmentation by utilizing natural language for class-specific embeddings. However, most research has focused on improving model accuracy through prompt engineering, prompt learning, or fine-tuning with limited labeled data, thereby overlooking the importance of refining the class descriptors. This paper introduces a new approach to text-supervised semantic segmentation using supervision by a large language model (LLM) that does not require extra training. …

abstract accuracy advanced arxiv class classification clip cs.cv data embeddings engineering fine-tuning free however importance improving language llm model accuracy natural natural language prompt prompt learning research segmentation semantic supervision through training type via zero-shot