Saved in:
Bibliographic Details
Main Authors: Yoon, Dongsik, Kim, Jongeun, Lee, Dayeon
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.10818
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910019005448192
author Yoon, Dongsik
Kim, Jongeun
Lee, Dayeon
author_facet Yoon, Dongsik
Kim, Jongeun
Lee, Dayeon
contents Action recognition on edge devices poses stringent constraints on latency, memory, storage, and power consumption. While auxiliary modalities such as skeleton and depth information can enhance recognition performance, they often require additional sensors or computationally expensive pose-estimation pipelines, limiting practicality for edge use. In this work, we propose a compact RGB-only network tailored for efficient on-device inference. Our approach builds upon an X3D-style backbone augmented with Temporal Shift, and further introduces selective temporal adaptation and parameter-free attention. Extensive experiments on the NTU RGB+D 60 and 120 benchmarks demonstrate a strong accuracy-efficiency balance. Moreover, deployment-level profiling on the Jetson Orin Nano verifies a smaller on-device footprint and practical resource utilization compared to existing RGB-based action recognition techniques.
format Preprint
id arxiv_https___arxiv_org_abs_2602_10818
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Resource-Efficient RGB-Only Action Recognition for Edge Deployment
Yoon, Dongsik
Kim, Jongeun
Lee, Dayeon
Computer Vision and Pattern Recognition
Performance
Action recognition on edge devices poses stringent constraints on latency, memory, storage, and power consumption. While auxiliary modalities such as skeleton and depth information can enhance recognition performance, they often require additional sensors or computationally expensive pose-estimation pipelines, limiting practicality for edge use. In this work, we propose a compact RGB-only network tailored for efficient on-device inference. Our approach builds upon an X3D-style backbone augmented with Temporal Shift, and further introduces selective temporal adaptation and parameter-free attention. Extensive experiments on the NTU RGB+D 60 and 120 benchmarks demonstrate a strong accuracy-efficiency balance. Moreover, deployment-level profiling on the Jetson Orin Nano verifies a smaller on-device footprint and practical resource utilization compared to existing RGB-based action recognition techniques.
title Resource-Efficient RGB-Only Action Recognition for Edge Deployment
topic Computer Vision and Pattern Recognition
Performance
url https://arxiv.org/abs/2602.10818