Graph-Based Multimodal and Multi-view Alignment for Keystep Recognition
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911435819319296 |
|---|---|
| author | Romero, Julia Lee Min, Kyle Tripathi, Subarna Karimzadeh, Morteza |
| author_facet | Romero, Julia Lee Min, Kyle Tripathi, Subarna Karimzadeh, Morteza |
| contents | Egocentric videos capture scenes from a wearer's viewpoint, resulting in dynamic backgrounds, frequent motion, and occlusions, posing challenges to accurate keystep recognition. We propose a flexible graph-learning framework for fine-grained keystep recognition that is able to effectively leverage long-term dependencies in egocentric videos, and leverage alignment between egocentric and exocentric videos during training for improved inference on egocentric videos. Our approach consists of constructing a graph where each video clip of the egocentric video corresponds to a node. During training, we consider each clip of each exocentric video (if available) as additional nodes. We examine several strategies to define connections across these nodes and pose keystep recognition as a node classification task on the constructed graphs. We perform extensive experiments on the Ego-Exo4D dataset and show that our proposed flexible graph-based framework notably outperforms existing methods by more than 12 points in accuracy. Furthermore, the constructed graphs are sparse and compute efficient. We also present a study examining on harnessing several multimodal features, including narrations, depth, and object class labels, on a heterogeneous graph and discuss their corresponding contribution to the keystep recognition performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_04121 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Graph-Based Multimodal and Multi-view Alignment for Keystep Recognition Romero, Julia Lee Min, Kyle Tripathi, Subarna Karimzadeh, Morteza Computer Vision and Pattern Recognition Egocentric videos capture scenes from a wearer's viewpoint, resulting in dynamic backgrounds, frequent motion, and occlusions, posing challenges to accurate keystep recognition. We propose a flexible graph-learning framework for fine-grained keystep recognition that is able to effectively leverage long-term dependencies in egocentric videos, and leverage alignment between egocentric and exocentric videos during training for improved inference on egocentric videos. Our approach consists of constructing a graph where each video clip of the egocentric video corresponds to a node. During training, we consider each clip of each exocentric video (if available) as additional nodes. We examine several strategies to define connections across these nodes and pose keystep recognition as a node classification task on the constructed graphs. We perform extensive experiments on the Ego-Exo4D dataset and show that our proposed flexible graph-based framework notably outperforms existing methods by more than 12 points in accuracy. Furthermore, the constructed graphs are sparse and compute efficient. We also present a study examining on harnessing several multimodal features, including narrations, depth, and object class labels, on a heterogeneous graph and discuss their corresponding contribution to the keystep recognition performance. |
| title | Graph-Based Multimodal and Multi-view Alignment for Keystep Recognition |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2501.04121 |