paper-with-me

홈 › Papers

Online Temporal Action Localization with Memory-Augmented Transformer

2024-08-06 · Youngkil Song, Dongkeun Kim, Minsu Cho, Suha Kwak

Online temporal action localization (On-TAL) is the task of identifying multiple action instances given a streaming video. Since existing methods take as input only a video segment of fixed size per iteration, they are limited in considering long-term context and require tuning the segment size carefully. To overcome these limitations, we propose memory-augmented transformer (MATR). MATR utilizes the memory queue that selectively preserves the past segment features, allowing to leverage long-term context for inference. We also propose a novel action localization method that observes the current input segment to predict the end time of the ongoing action and accesses the memory queue to estimate the start time of the action. Our method outperformed existing methods on two datasets, THUMOS14 and MUSES, surpassing not only TAL methods in the online setting but also some offline TAL methods.

📄 PDF Abstract BibTeX arXiv:2408.02957

Code (0)

등록된 구현이 없습니다.

Tasks

Action LocalizationTemporal Action Localization

Similar Papers 제목 키워드 기반

HAT: History-Augmented Anchor Transformer for Online Temporal Action Localization

2024-08-12 · Sakib Reza, Yuexi Zhang, Mohsen Moghaddam, Octavia Camps

Online video understanding often relies on individual frames, leading to frame-by-frame predictions. Recent advancements such as Online Temporal Action Localization (OnTAL), extend this approach to instance-level predict…

Action LocalizationTemporal Action LocalizationVideo Understanding

TALLFormer: Temporal Action Localization with a Long-memory Transformer

2022-04-04 · Feng Cheng, Gedas Bertasius

Most modern approaches in temporal action localization divide this problem into two parts: (i) short-term feature extraction and (ii) long-range temporal boundary localization. Due to the high GPU memory cost caused by p…

Action LocalizationAction RecognitionGPUTemporal Action Localization

Action Unit Memory Network for Weakly Supervised Temporal Action Localization

2021-04-29 · CVPR 2021 1 · Wang Luo, Tianzhu Zhang, Wenfei Yang, Jingen Liu 외

Weakly supervised temporal action localization aims to detect and localize actions in untrimmed videos with only video-level labels during training. However, without frame-level annotations, it is challenging to achieve …

Action LocalizationDiversityTemporal Action LocalizationWeakly Supervised Action Localization+1

Weakly-Supervised Temporal Action Localization with Bidirectional Semantic Consistency Constraint

2023-04-25 · Guozhang Li, De Cheng, Xinpeng Ding, Nannan Wang 외

Weakly Supervised Temporal Action Localization (WTAL) aims to classify and localize temporal boundaries of actions for the video, given only video-level category labels in the training datasets. Due to the lack of bounda…

Action LocalizationTemporal Action LocalizationWeakly-supervised Temporal Action Localization

Progress Regression RNN for Online Spatial-Temporal Action Localization in Unconstrained Videos

2019-03-01 · Bo Hu, Jianfei Cai, Tat-Jen Cham, Junsong Yuan

Previous spatial-temporal action localization methods commonly follow the pipeline of object detection to estimate bounding boxes and labels of actions. However, the temporal relation of an action has not been fully expl…

Action Localizationobject-detectionObject DetectionPosition+2