GRITv2: Efficient and Light-weight Social Relation Recognition

arXiv:2403.06895v1 Announce Type: new
Abstract: Our research focuses on the analysis and improvement of the Graph-based Relation Inference Transformer (GRIT), which serves as an important benchmark in the field. We conduct a comprehensive ablation study using the PISC-fine dataset, to find and explore improvement in efficiency and performance of GRITv2. Our research has provided a new state-of-the-art relation recognition model on the PISC relation dataset. We introduce several features in the GRIT model and analyse our new benchmarks in two …

abstract analysis arxiv benchmark cs.cv dataset efficiency explore graph graph-based grit improvement inference light performance recognition research social study transformer type