Temporal Context Consistency Above All: Enhancing Long-Term Anticipation by Learning and Enforcing Temporal Constraints

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Maté, Alberto, Dimiccoli, Mariella
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913627594817536
author Maté, Alberto
Dimiccoli, Mariella
author_facet Maté, Alberto
Dimiccoli, Mariella
contents This paper proposes a method for long-term action anticipation (LTA), the task of predicting action labels and their duration in a video given the observation of an initial untrimmed video interval. We build on an encoder-decoder architecture with parallel decoding and make two key contributions. First, we introduce a bi-directional action context regularizer module on the top of the decoder that ensures temporal context coherence in temporally adjacent segments. Second, we learn from classified segments a transition matrix that models the probability of transitioning from one action to another and the sequence is optimized globally over the full prediction interval. In addition, we use a specialized encoder for the task of action segmentation to increase the quality of the predictions in the observation interval at inference time, leading to a better understanding of the past. We validate our methods on four benchmark datasets for LTA, the EpicKitchen-55, EGTEA+, 50Salads and Breakfast demonstrating superior or comparable performance to state-of-the-art methods, including probabilistic models and also those based on Large Language Models, that assume trimmed video as input. The code will be released upon acceptance.
format Preprint
id arxiv_https___arxiv_org_abs_2412_19424
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Temporal Context Consistency Above All: Enhancing Long-Term Anticipation by Learning and Enforcing Temporal Constraints
Maté, Alberto
Dimiccoli, Mariella
Computer Vision and Pattern Recognition
This paper proposes a method for long-term action anticipation (LTA), the task of predicting action labels and their duration in a video given the observation of an initial untrimmed video interval. We build on an encoder-decoder architecture with parallel decoding and make two key contributions. First, we introduce a bi-directional action context regularizer module on the top of the decoder that ensures temporal context coherence in temporally adjacent segments. Second, we learn from classified segments a transition matrix that models the probability of transitioning from one action to another and the sequence is optimized globally over the full prediction interval. In addition, we use a specialized encoder for the task of action segmentation to increase the quality of the predictions in the observation interval at inference time, leading to a better understanding of the past. We validate our methods on four benchmark datasets for LTA, the EpicKitchen-55, EGTEA+, 50Salads and Breakfast demonstrating superior or comparable performance to state-of-the-art methods, including probabilistic models and also those based on Large Language Models, that assume trimmed video as input. The code will be released upon acceptance.
title Temporal Context Consistency Above All: Enhancing Long-Term Anticipation by Learning and Enforcing Temporal Constraints
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.19424