Spatio-Temporal Dynamic Inference Network for Group Activity Recognition

By Hangjie Yuan, Dong Ni, Mang Wang. 2021

116 citations recorded by the source corpus.

Group activity recognition aims to understand the activity performed by a group of people. In order to solve it, modeling complex spatiotemporal interactions is the key. Previous methods are limited in reasoning on a predefined graph, which ignores the inherent person-specific interaction context. Moreover, they adopt inference schemes that are computationally expensive and easily result in the over-smoothing problem. In this paper, we manage to achieve spatio-temporal person-specific inferences by proposing Dynamic Inference Network (DIN), which composes of Dynamic Relation (DR) module and Dynamic Walk (DW) module. We firstly propose to initialize interaction fields on a primary spatio-temporal graph. Within each interaction field, we apply DR to predict the relation matrix and DW to predict the dynamic walk offsets in a joint-processing manner, thus forming a person-specific interaction graph. By updating features on the specific graph, a person can possess a global-level interaction field with a local initialization. Experiments indicate both modules’ effectiveness. Moreover, DIN1 achieves significant improvement compared to previous state-of-the-art methods on two popular datasets under the same setting, while costing much less computation overhead of the reasoning module.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Mang Wang
2. Hangjie Yuan
3. Liyuan Wang

Citations in the corpus

• Overcoming Catastrophic Forgetting in Incremental Object Detection via Elastic Response Distillation (2022, background): “This learning paradigm implicitly assumes data distribution is fixed or stationary [9, 37], while data from real world is represented by continuous”