English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The task of Group Activity Recognition (GAR) aims to predict the activitycategory of the group by learning the actor spatial-temporal interactionrelation in the group. Therefore, an effective actor relation learning methodis crucial for the GAR task. The previous works mainly learn the interactionrelation by the well-designed GCNs or Transformers. For example, to infer theactor interaction relation, GCNs need a learnable adjacency, and Transformersneed to calculate the self-attention. Although the above methods can model theinteraction relation effectively, they also increase the complexity of themodel (the number of parameters and computations). In this paper, we design anovel MLP-based method for Actor Interaction Relation learning (MLP-AIR) inGAR. Compared with GCNs and Transformers, our method has a competitive butconceptually and technically simple alternative, significantly reducing thecomplexity. Specifically, MLP-AIR includes three sub-modules: MLP-based Spatialrelation modeling module (MLP-S), MLP-based Temporal relation modeling module(MLP-T), and MLP-based Relation refining module (MLP-R). MLP-S is used to modelthe spatial relation between different actors in each frame. MLP-T is used tomodel the temporal relation between different frames for each actor. MLP-R isused further to refine the relation between different dimensions of relationfeatures to improve the feature's expression ability. To evaluate the MLP-AIR,we conduct extensive experiments on two widely used benchmarks, including theVolleyball and Collective Activity datasets. Experimental results demonstratethat MLP-AIR can get competitive results but with low complexity.

MLP-AIR: An Efficient MLP-Based Method for Actor Interaction Relation Learning in Group Activity Recognition