Related papers: Towards Open-vocabulary Scene Graph Generation with Prompt-based Finetuning

Towards Open-vocabulary Scene Graph Generation with Prompt-based Finetuning

URL: http://arxiv.org/abs/2208.08165v1
Date: Wed, 17 Aug 2022 09:05:38 GMT
Title: Towards Open-vocabulary Scene Graph Generation with Prompt-based Finetuning
Authors: Tao He, Lianli Gao, Jingkuan Song, Yuan-Fang Li
Abstract summary: Scene graph generation (SGG) is a fundamental task aimed at detecting visual relations between objects in an image. We introduce open-vocabulary scene graph generation, a novel, realistic and challenging setting in which a model is trained on a set of base object classes. Our method can support inference over completely unseen object classes, which existing methods are incapable of handling.
Score: 84.39787427288525
License: http://creativecommons.org/licenses/by/4.0/
Abstract: Scene graph generation (SGG) is a fundamental task aimed at detecting visual relations between objects in an image. The prevailing SGG methods require all object classes to be given in the training set. Such a closed setting limits the practical application of SGG. In this paper, we introduce open-vocabulary scene graph generation, a novel, realistic and challenging setting in which a model is trained on a set of base object classes but is required to infer relations for unseen target object classes. To this end, we propose a two-step method that firstly pre-trains on large amounts of coarse-grained region-caption data and then leverages two prompt-based techniques to finetune the pre-trained model without updating its parameters. Moreover, our method can support inference over completely unseen object classes, which existing methods are incapable of handling. On extensive experiments on three benchmark datasets, Visual Genome, GQA, and Open-Image, our method significantly outperforms recent, strong SGG methods on the setting of Ov-SGG, as well as on the conventional closed SGG.