Context-Enhanced Memory-Refined Transformer for Online Action Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pang, Zhanzhong, Sener, Fadime, Yao, Angela
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908280493703168
author Pang, Zhanzhong
Sener, Fadime
Yao, Angela
author_facet Pang, Zhanzhong
Sener, Fadime
Yao, Angela
contents Online Action Detection (OAD) detects actions in streaming videos using past observations. State-of-the-art OAD approaches model past observations and their interactions with an anticipated future. The past is encoded using short- and long-term memories to capture immediate and long-range dependencies, while anticipation compensates for missing future context. We identify a training-inference discrepancy in existing OAD methods that hinders learning effectiveness. The training uses varying lengths of short-term memory, while inference relies on a full-length short-term memory. As a remedy, we propose a Context-enhanced Memory-Refined Transformer (CMeRT). CMeRT introduces a context-enhanced encoder to improve frame representations using additional near-past context. It also features a memory-refined decoder to leverage near-future generation to enhance performance. CMeRT achieves state-of-the-art in online detection and anticipation on THUMOS'14, CrossTask, and EPIC-Kitchens-100.
format Preprint
id arxiv_https___arxiv_org_abs_2503_18359
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Context-Enhanced Memory-Refined Transformer for Online Action Detection
Pang, Zhanzhong
Sener, Fadime
Yao, Angela
Computer Vision and Pattern Recognition
Online Action Detection (OAD) detects actions in streaming videos using past observations. State-of-the-art OAD approaches model past observations and their interactions with an anticipated future. The past is encoded using short- and long-term memories to capture immediate and long-range dependencies, while anticipation compensates for missing future context. We identify a training-inference discrepancy in existing OAD methods that hinders learning effectiveness. The training uses varying lengths of short-term memory, while inference relies on a full-length short-term memory. As a remedy, we propose a Context-enhanced Memory-Refined Transformer (CMeRT). CMeRT introduces a context-enhanced encoder to improve frame representations using additional near-past context. It also features a memory-refined decoder to leverage near-future generation to enhance performance. CMeRT achieves state-of-the-art in online detection and anticipation on THUMOS'14, CrossTask, and EPIC-Kitchens-100.
title Context-Enhanced Memory-Refined Transformer for Online Action Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.18359