Spectral Scalpel: Amplifying Adjacent Action Discrepancy via Frequency-Selective Filtering for Skeleton-Based Action Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ji, Haoyu, Chen, Bowen, Yang, Zhihao, Huang, Wenze, Gao, Yu, Liu, Xueting, Ren, Weihong, Wang, Zhiyong, Liu, Honghai
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915889700405248
author Ji, Haoyu
Chen, Bowen
Yang, Zhihao
Huang, Wenze
Gao, Yu
Liu, Xueting
Ren, Weihong
Wang, Zhiyong
Liu, Honghai
author_facet Ji, Haoyu
Chen, Bowen
Yang, Zhihao
Huang, Wenze
Gao, Yu
Liu, Xueting
Ren, Weihong
Wang, Zhiyong
Liu, Honghai
contents Skeleton-based Temporal Action Segmentation (STAS) seeks to densely segment and classify diverse actions within long, untrimmed skeletal motion sequences. However, existing STAS methodologies face challenges of limited inter-class discriminability and blurred segmentation boundaries, primarily due to insufficient distinction of spatio-temporal patterns between adjacent actions. To address these limitations, we propose Spectral Scalpel, a frequency-selective filtering framework aimed at suppressing shared frequency components between adjacent distinct actions while amplifying their action-specific frequencies, thereby enhancing inter-action discrepancies and sharpening transition boundaries. Specifically, Spectral Scalpel employs adaptive multi-scale spectral filters as scalpels to edit frequency spectra, coupled with a discrepancy loss between adjacent actions serving as the surgical objective. This design amplifies representational disparities between neighboring actions, effectively mitigating boundary localization ambiguities and inter-class confusion. Furthermore, complementing long-term temporal modeling, we introduce a frequency-aware channel mixer to strengthen channel evolution by aggregating spectra across channels. This work presents a novel paradigm for STAS that extends conventional spatio-temporal modeling by incorporating frequency-domain analysis. Extensive experiments on five public datasets demonstrate that Spectral Scalpel achieves state-of-the-art performance. Code is available at https://github.com/HaoyuJi/SpecScalpel.
format Preprint
id arxiv_https___arxiv_org_abs_2603_24134
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Spectral Scalpel: Amplifying Adjacent Action Discrepancy via Frequency-Selective Filtering for Skeleton-Based Action Segmentation
Ji, Haoyu
Chen, Bowen
Yang, Zhihao
Huang, Wenze
Gao, Yu
Liu, Xueting
Ren, Weihong
Wang, Zhiyong
Liu, Honghai
Computer Vision and Pattern Recognition
Skeleton-based Temporal Action Segmentation (STAS) seeks to densely segment and classify diverse actions within long, untrimmed skeletal motion sequences. However, existing STAS methodologies face challenges of limited inter-class discriminability and blurred segmentation boundaries, primarily due to insufficient distinction of spatio-temporal patterns between adjacent actions. To address these limitations, we propose Spectral Scalpel, a frequency-selective filtering framework aimed at suppressing shared frequency components between adjacent distinct actions while amplifying their action-specific frequencies, thereby enhancing inter-action discrepancies and sharpening transition boundaries. Specifically, Spectral Scalpel employs adaptive multi-scale spectral filters as scalpels to edit frequency spectra, coupled with a discrepancy loss between adjacent actions serving as the surgical objective. This design amplifies representational disparities between neighboring actions, effectively mitigating boundary localization ambiguities and inter-class confusion. Furthermore, complementing long-term temporal modeling, we introduce a frequency-aware channel mixer to strengthen channel evolution by aggregating spectra across channels. This work presents a novel paradigm for STAS that extends conventional spatio-temporal modeling by incorporating frequency-domain analysis. Extensive experiments on five public datasets demonstrate that Spectral Scalpel achieves state-of-the-art performance. Code is available at https://github.com/HaoyuJi/SpecScalpel.
title Spectral Scalpel: Amplifying Adjacent Action Discrepancy via Frequency-Selective Filtering for Skeleton-Based Action Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.24134