TaskCLIP: Extend Large Vision-Language Model for Task Oriented Object Detection

arXiv:2403.08108v1 Announce Type: new
Abstract: Task-oriented object detection aims to find objects suitable for accomplishing specific tasks. As a challenging task, it requires simultaneous visual data processing and reasoning under ambiguous semantics. Recent solutions are mainly all-in-one models. However, the object detection backbones are pre-trained without text supervision. Thus, to incorporate task requirements, their intricate models undergo extensive learning on a highly imbalanced and scarce dataset, resulting in capped performance, laborious training, and poor generalizability. In contrast, we propose TaskCLIP, …

abstract arxiv cs.cv data data processing detection however language language model object objects processing reasoning semantics solutions specific tasks supervision tasks text type vision visual visual data