CoCoDiff: Diversifying Skeleton Action Features via Coarse-Fine Text-Co-Guided Latent Diffusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhao, Zhifu, Hua, Hanyang, Li, Jianan, Wu, Shaoxin, Li, Fu, Zhou, Yangtao, Li, Yang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913812916994048
author Zhao, Zhifu
Hua, Hanyang
Li, Jianan
Wu, Shaoxin
Li, Fu
Zhou, Yangtao
Li, Yang
author_facet Zhao, Zhifu
Hua, Hanyang
Li, Jianan
Wu, Shaoxin
Li, Fu
Zhou, Yangtao
Li, Yang
contents In action recognition tasks, feature diversity is essential for enhancing model generalization and performance. Existing methods typically promote feature diversity by expanding the training data in the sample space, which often leads to inefficiencies and semantic inconsistencies. To overcome these problems, we propose a novel Coarse-fine text co-guidance Diffusion model (CoCoDiff). CoCoDiff generates diverse yet semantically consistent features in the latent space by leveraging diffusion and multi-granularity textual guidance. Specifically, our approach feeds spatio-temporal features extracted from skeleton sequences into a latent diffusion model to generate diverse action representations. Meanwhile, we introduce a coarse-fine text co-guided strategy that leverages textual information from large language models (LLMs) to ensure semantic consistency between the generated features and the original inputs. It is noted that CoCoDiff operates as a plug-and-play auxiliary module during training, incurring no additional inference cost. Extensive experiments demonstrate that CoCoDiff achieves SOTA performance on skeleton-based action recognition benchmarks, including NTU RGB+D, NTU RGB+D 120 and Kinetics-Skeleton.
format Preprint
id arxiv_https___arxiv_org_abs_2504_21266
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CoCoDiff: Diversifying Skeleton Action Features via Coarse-Fine Text-Co-Guided Latent Diffusion
Zhao, Zhifu
Hua, Hanyang
Li, Jianan
Wu, Shaoxin
Li, Fu
Zhou, Yangtao
Li, Yang
Computer Vision and Pattern Recognition
In action recognition tasks, feature diversity is essential for enhancing model generalization and performance. Existing methods typically promote feature diversity by expanding the training data in the sample space, which often leads to inefficiencies and semantic inconsistencies. To overcome these problems, we propose a novel Coarse-fine text co-guidance Diffusion model (CoCoDiff). CoCoDiff generates diverse yet semantically consistent features in the latent space by leveraging diffusion and multi-granularity textual guidance. Specifically, our approach feeds spatio-temporal features extracted from skeleton sequences into a latent diffusion model to generate diverse action representations. Meanwhile, we introduce a coarse-fine text co-guided strategy that leverages textual information from large language models (LLMs) to ensure semantic consistency between the generated features and the original inputs. It is noted that CoCoDiff operates as a plug-and-play auxiliary module during training, incurring no additional inference cost. Extensive experiments demonstrate that CoCoDiff achieves SOTA performance on skeleton-based action recognition benchmarks, including NTU RGB+D, NTU RGB+D 120 and Kinetics-Skeleton.
title CoCoDiff: Diversifying Skeleton Action Features via Coarse-Fine Text-Co-Guided Latent Diffusion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.21266