STMixer: A One-Stage Sparse Action Detector

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Tao, Cao, Mengqi, Gao, Ziteng, Wu, Gangshan, Wang, Limin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916206646132736
author Wu, Tao
Cao, Mengqi
Gao, Ziteng
Wu, Gangshan
Wang, Limin
author_facet Wu, Tao
Cao, Mengqi
Gao, Ziteng
Wu, Gangshan
Wang, Limin
contents Traditional video action detectors typically adopt the two-stage pipeline, where a person detector is first employed to generate actor boxes and then 3D RoIAlign is used to extract actor-specific features for classification. This detection paradigm requires multi-stage training and inference, and the feature sampling is constrained inside the box, failing to effectively leverage richer context information outside. Recently, a few query-based action detectors have been proposed to predict action instances in an end-to-end manner. However, they still lack adaptability in feature sampling and decoding, thus suffering from the issues of inferior performance or slower convergence. In this paper, we propose two core designs for a more flexible one-stage sparse action detector. First, we present a query-based adaptive feature sampling module, which endows the detector with the flexibility of mining a group of discriminative features from the entire spatio-temporal domain. Second, we devise a decoupled feature mixing module, which dynamically attends to and mixes video features along the spatial and temporal dimensions respectively for better feature decoding. Based on these designs, we instantiate two detection pipelines, that is, STMixer-K for keyframe action detection and STMixer-T for action tubelet detection. Without bells and whistles, our STMixer detectors obtain state-of-the-art results on five challenging spatio-temporal action detection benchmarks for keyframe action detection or action tube detection.
format Preprint
id arxiv_https___arxiv_org_abs_2404_09842
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle STMixer: A One-Stage Sparse Action Detector
Wu, Tao
Cao, Mengqi
Gao, Ziteng
Wu, Gangshan
Wang, Limin
Computer Vision and Pattern Recognition
Traditional video action detectors typically adopt the two-stage pipeline, where a person detector is first employed to generate actor boxes and then 3D RoIAlign is used to extract actor-specific features for classification. This detection paradigm requires multi-stage training and inference, and the feature sampling is constrained inside the box, failing to effectively leverage richer context information outside. Recently, a few query-based action detectors have been proposed to predict action instances in an end-to-end manner. However, they still lack adaptability in feature sampling and decoding, thus suffering from the issues of inferior performance or slower convergence. In this paper, we propose two core designs for a more flexible one-stage sparse action detector. First, we present a query-based adaptive feature sampling module, which endows the detector with the flexibility of mining a group of discriminative features from the entire spatio-temporal domain. Second, we devise a decoupled feature mixing module, which dynamically attends to and mixes video features along the spatial and temporal dimensions respectively for better feature decoding. Based on these designs, we instantiate two detection pipelines, that is, STMixer-K for keyframe action detection and STMixer-T for action tubelet detection. Without bells and whistles, our STMixer detectors obtain state-of-the-art results on five challenging spatio-temporal action detection benchmarks for keyframe action detection or action tube detection.
title STMixer: A One-Stage Sparse Action Detector
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2404.09842