# Efficient Egocentric Action Recognition with Multimodal Data

arXiv: https://arxiv.org/abs/2506.01757 · HTML: https://mlsanity.com/paper/2506.01757/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 18 references that are in the map.
- First posted on arXiv 2 Jun 2025.

## Builds on (most cited first, 13 of 13)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://mlsanity.com/paper/1506.04214/) (2015)
- [Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition](https://mlsanity.com/paper/1801.07455/) (2018)
- [Two-Stream Convolutional Networks for Action Recognition in Videos](https://mlsanity.com/paper/1406.2199/) (2014)
- [Designing Network Design Spaces](https://mlsanity.com/paper/2003.13678/) (2020)
- [LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inference](https://mlsanity.com/paper/2104.01136/) (2021)
- [H2O: Two Hands Manipulating Objects for First Person Interaction Recognition](https://mlsanity.com/paper/2104.11181/) (2021)
- [HoloAssist: an Egocentric Human Interaction Dataset for Interactive AI Assistants in the Real World](https://mlsanity.com/paper/2309.17024/) (2023)
- [Avatars Grow Legs: Generating Smooth Human Motion from Sparse Tracking Inputs with Diffusion Model](https://mlsanity.com/paper/2304.08577/) (2023)
- [What Would You Expect? Anticipating Egocentric Actions with Rolling-Unrolling LSTMs and Modality Attention](https://mlsanity.com/paper/1905.09035/) (2019)
- [Attention is All We Need: Nailing Down Object-centric Attention for Egocentric Activity Recognition](https://mlsanity.com/paper/1807.11794/) (2018)
- [On the Utility of 3D Hand Poses for Action Recognition](https://mlsanity.com/paper/2403.09805/) (2024)
- [Convolutional Long Short-Term Memory Networks for Recognizing First Person Interactions](https://mlsanity.com/paper/1709.06495/) (2017)

## Abstract (arXiv)

The increasing availability of wearable XR devices opens new perspectives for Egocentric Action Recognition (EAR) systems, which can provide deeper human understanding and situation awareness. However, deploying real-time algorithms on these devices can be challenging due to the inherent trade-offs between portability, battery life, and computational resources. In this work, we systematically analyze the impact of sampling frequency across different input modalities - RGB video and 3D hand pose - on egocentric action recognition performance and CPU usage. By exploring a range of configurations, we provide a comprehensive characterization of the trade-offs between accuracy and computational efficiency. Our findings reveal that reducing the sampling rate of RGB frames, when complemented with higher-frequency 3D hand pose input, can preserve high accuracy while significantly lowering CPU demands. Notably, we observe up to a 3x reduction in CPU usage with minimal to no loss in recognition performance. This highlights the potential of multimodal input strategies as a viable approach to achieving efficient, real-time EAR on XR devices.