Jan 28, 2024 · In this paper, we propose a novel Parallel Adaptive Graph-Based Network (PAGN) to balance conflicts between long and short actions.

This results in feature outputs that can represent precise location information and rich semantic information simultaneously, making it more efficient and ...

To deal with these problems, we propose the Parallel Adaptive Graph-Based Network (PAGN), which constructs a multi-branch parallel subnetwork that retains ...

Learning Complementary Instance Representation with Parallel Adaptive Graph-Based Network for Action Detection. https://doi.org/10.1007/978-3-031-53308-2_34 ...

Learning Complementary Instance Representation with Parallel Adaptive Graph-Based Network for Action Detection. Y. Jiao, W. Yang, and W. Xing.

A general framework for graph-level clustering, scaling up dynamic graph representation learning via spiking neural networks, generalizing downsampling from ...

We propose an action detection network using temporal feature pyramid, which can collect data using cameras and predict precise action categories and ...

The dataset used in this article is NTU-RGB-60, 120 and Kinetics for skeleton based human action recognition. ResearchGate Logo. Discover the world's research.

Apr 28, 2021 · We present an efficient approach for temporal action co-localization (TACL), which means to simultaneously localize all action instances in an untrimmed video.

We collect existing papers on skeleton-based action recognition published in prominent conferences and journals. This paper list will be continuously updated.