Permutation-Aware Action Segmentation via Unsupervised Frame-to-Segment Alignment

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tran, Quoc-Huy, Mehmood, Ahmed, Ahmed, Muhammad, Naufil, Muhammad, Zafar, Anas, Konin, Andrey, Zia, M. Zeeshan
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908684825657344
author Tran, Quoc-Huy
Mehmood, Ahmed
Ahmed, Muhammad
Naufil, Muhammad
Zafar, Anas
Konin, Andrey
Zia, M. Zeeshan
author_facet Tran, Quoc-Huy
Mehmood, Ahmed
Ahmed, Muhammad
Naufil, Muhammad
Zafar, Anas
Konin, Andrey
Zia, M. Zeeshan
contents This paper presents an unsupervised transformer-based framework for temporal activity segmentation which leverages not only frame-level cues but also segment-level cues. This is in contrast with previous methods which often rely on frame-level information only. Our approach begins with a frame-level prediction module which estimates framewise action classes via a transformer encoder. The frame-level prediction module is trained in an unsupervised manner via temporal optimal transport. To exploit segment-level information, we utilize a segment-level prediction module and a frame-to-segment alignment module. The former includes a transformer decoder for estimating video transcripts, while the latter matches frame-level features with segment-level features, yielding permutation-aware segmentation results. Moreover, inspired by temporal optimal transport, we introduce simple-yet-effective pseudo labels for unsupervised training of the above modules. Our experiments on four public datasets, i.e., 50 Salads, YouTube Instructions, Breakfast, and Desktop Assembly show that our approach achieves comparable or better performance than previous methods in unsupervised activity segmentation. Our code and dataset are available on our research website: https://retrocausal.ai/research/.
format Preprint
id arxiv_https___arxiv_org_abs_2305_19478
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Permutation-Aware Action Segmentation via Unsupervised Frame-to-Segment Alignment
Tran, Quoc-Huy
Mehmood, Ahmed
Ahmed, Muhammad
Naufil, Muhammad
Zafar, Anas
Konin, Andrey
Zia, M. Zeeshan
Computer Vision and Pattern Recognition
This paper presents an unsupervised transformer-based framework for temporal activity segmentation which leverages not only frame-level cues but also segment-level cues. This is in contrast with previous methods which often rely on frame-level information only. Our approach begins with a frame-level prediction module which estimates framewise action classes via a transformer encoder. The frame-level prediction module is trained in an unsupervised manner via temporal optimal transport. To exploit segment-level information, we utilize a segment-level prediction module and a frame-to-segment alignment module. The former includes a transformer decoder for estimating video transcripts, while the latter matches frame-level features with segment-level features, yielding permutation-aware segmentation results. Moreover, inspired by temporal optimal transport, we introduce simple-yet-effective pseudo labels for unsupervised training of the above modules. Our experiments on four public datasets, i.e., 50 Salads, YouTube Instructions, Breakfast, and Desktop Assembly show that our approach achieves comparable or better performance than previous methods in unsupervised activity segmentation. Our code and dataset are available on our research website: https://retrocausal.ai/research/.
title Permutation-Aware Action Segmentation via Unsupervised Frame-to-Segment Alignment
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2305.19478