The Solution for the CVPR 2023 1st foundation model challenge-Track2

arXiv:2403.17702v1 Announce Type: new
Abstract: In this paper, we propose a solution for cross-modal transportation retrieval. Due to the cross-domain problem of traffic images, we divide the problem into two sub-tasks of pedestrian retrieval and vehicle retrieval through a simple strategy. In pedestrian retrieval tasks, we use IRRA as the base model and specifically design an Attribute Classification to mine the knowledge implied by attribute labels. More importantly, We use the strategy of Inclusion Relation Matching to make the image-text …

abstract arxiv challenge cs.cv cvpr domain foundation foundation model images modal paper pedestrian retrieval simple solution strategy tasks through traffic transportation type