Learning from Semantic Alignment between Unpaired Multiviews for Egocentric Video Recognition

TL;DR

This paper introduces SUM-L, a novel framework for unpaired multiview egocentric video recognition that leverages semantic alignment and cross-view pseudo-pairs to improve view-invariant representations.

Contribution

The paper proposes a new semantic-based unpaired multiview learning method that aligns views and videos using semantic information, outperforming existing methods in challenging unpaired scenarios.

Findings

Outperforms existing view-alignment methods on benchmark datasets.

Effectively leverages semantic knowledge through video-text alignment.

Demonstrates robustness in unpaired multiview egocentric video recognition.

Abstract

We are concerned with a challenging scenario in unpaired multiview video learning. In this case, the model aims to learn comprehensive multiview representations while the cross-view semantic information exhibits variations. We propose Semantics-based Unpaired Multiview Learning (SUM-L) to tackle this unpaired multiview learning problem. The key idea is to build cross-view pseudo-pairs and do view-invariant alignment by leveraging the semantic information of videos. To facilitate the data efficiency of multiview learning, we further perform video-text alignment for first-person and third-person videos, to fully leverage the semantic knowledge to improve video representations. Extensive experiments on multiple benchmark datasets verify the effectiveness of our framework. Our method also outperforms multiple existing view-alignment methods, under the more challenging scenario than typical paired or unpaired multimodal or multiview learning. Our code is available at https://github.com/wqtwjt1996/SUM-L.