# Temporal Reasoning Graph for Activity Recognition

arXiv: https://arxiv.org/abs/1908.09995 · HTML: https://mlsanity.com/paper/1908.09995/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Aug 2019.

## Built on by (most cited first, 22 of 22)

- [ACTION-Net: Multipath Excitation for Action Recognition](https://mlsanity.com/paper/2103.07372/) (2021)
- [A Survey on Graph Neural Networks and Graph Transformers in Computer Vision: A Task-Oriented Perspective](https://mlsanity.com/paper/2209.13232/) (2022)
- [Multiscale Spatio-Temporal Graph Neural Networks for 3D Skeleton-Based Motion Prediction](https://mlsanity.com/paper/2108.11244/) (2021)
- [TCGL: Temporal Contrastive Graph for Self-supervised Video Representation Learning](https://mlsanity.com/paper/2112.03587/) (2021)
- [DirecFormer: A Directed Attention in Transformer Approach to Robust Action Recognition](https://mlsanity.com/paper/2203.10233/) (2022)
- [Skeleton-Parted Graph Scattering Networks for 3D Human Motion Prediction](https://mlsanity.com/paper/2208.00368/) (2022)
- [SOLVER: Scene-Object Interrelated Visual Emotion Reasoning Network](https://mlsanity.com/paper/2110.12334/) (2021)
- [Long-Term Rhythmic Video Soundtracker](https://mlsanity.com/paper/2305.01319/) (2023)
- [Multi-Modal Interaction Graph Convolutional Network for Temporal Language Localization in Videos](https://mlsanity.com/paper/2110.06058/) (2021)
- [Grounding Partially-Defined Events in Multimodal Data](https://mlsanity.com/paper/2410.05267/) (2024)
- [Representing Videos as Discriminative Sub-graphs for Action Recognition](https://mlsanity.com/paper/2201.04027/) (2022)
- [JOADAA: joint online action detection and action anticipation](https://mlsanity.com/paper/2309.06130/) (2023)
- [Temporal Contrastive Graph Learning for Video Action Recognition and Retrieval](https://mlsanity.com/paper/2101.00820/) (2021)
- [MECD+: Unlocking Event-Level Causal Graph Discovery for Video Reasoning](https://mlsanity.com/paper/2501.07227/) (2025)
- [Video Is Graph: Structured Graph Module for Video Action Recognition](https://mlsanity.com/paper/2110.05904/) (2021)
- [Human-like Relational Models for Activity Recognition in Video](https://mlsanity.com/paper/2107.05319/) (2021)
- [Prompt-guided Disentangled Representation for Action Recognition](https://mlsanity.com/paper/2509.21783/) (2025)
- [HAR-GCNN: Deep Graph CNNs for Human Activity Recognition From Highly Unlabeled Mobile Sensor Data](https://mlsanity.com/paper/2203.03087/) (2022)
- [Data Transformation Strategies to Remove Heterogeneity](https://mlsanity.com/paper/2507.12677/) (2025)
- [Knowledge-guided Disentanglement with Atomic Actions for Action Recognition](https://mlsanity.com/paper/2607.26097/) (2026)
- [G3Ego: Gaze-Guided Graphs for Egocentric Action Understanding](https://mlsanity.com/paper/2608.20157/) (2026)
- [TReViS: Temporal Repetition Structure Aware Video Synthesis for Self-supervised Repetitive Action Counting](https://mlsanity.com/paper/2609.24367/) (2026)

## Abstract (arXiv)

Despite great success has been achieved in activity analysis, it still has many challenges. Most existing work in activity recognition pay more attention to design efficient architecture or video sampling strategy. However, due to the property of fine-grained action and long term structure in video, activity recognition is expected to reason temporal relation between video sequences. In this paper, we propose an efficient temporal reasoning graph (TRG) to simultaneously capture the appearance features and temporal relation between video sequences at multiple time scales. Specifically, we construct learnable temporal relation graphs to explore temporal relation on the multi-scale range. Additionally, to facilitate multi-scale temporal relation extraction, we design a multi-head temporal adjacent matrix to represent multi-kinds of temporal relations. Eventually, a multi-head temporal relation aggregator is proposed to extract the semantic meaning of those features convolving through the graphs. Extensive experiments are performed on widely-used large-scale datasets, such as Something-Something and Charades, and the results show that our model can achieve state-of-the-art performance. Further analysis shows that temporal relation reasoning with our TRG can extract discriminative features for activity recognition.