Cross-Model Cross-Stream Learning for Self-Supervised Human Action Recognition

TL;DR

This paper introduces a novel self-supervised learning framework for skeleton-based human action recognition that leverages cross-model and cross-stream strategies to improve feature discrimination and multi-stream information integration.

Contribution

It proposes the CMCS framework combining CMAL and CSCL, and introduces SkeletonBYOL as a baseline for effective self-supervised skeleton action recognition.

Findings

Outperforms state-of-the-art methods on three datasets.

Demonstrates the effectiveness of cross-model adversarial and collaborative learning.

Shows improved discriminative feature learning within single streams.

Abstract

Considering the instance-level discriminative ability, contrastive learning methods, including MoCo and SimCLR, have been adapted from the original image representation learning task to solve the self-supervised skeleton-based action recognition task. These methods usually use multiple data streams (i.e., joint, motion, and bone) for ensemble learning, meanwhile, how to construct a discriminative feature space within a single stream and effectively aggregate the information from multiple streams remains an open problem. To this end, this paper first applies a new contrastive learning method called BYOL to learn from skeleton data, and then formulate SkeletonBYOL as a simple yet effective baseline for self-supervised skeleton-based action recognition. Inspired by SkeletonBYOL, this paper further presents a Cross-Model and Cross-Stream (CMCS) framework. This framework combines Cross-Model Adversarial Learning (CMAL) and Cross-Stream Collaborative Learning (CSCL). Specifically, CMAL learns single-stream representation by cross-model adversarial loss to obtain more discriminative features. To aggregate and interact with multi-stream information, CSCL is designed by generating similarity pseudo label of ensemble learning as supervision and guiding feature generation for individual streams. Extensive experiments on three datasets verify the complementary properties between CMAL and CSCL and also verify that the proposed method can achieve better results than state-of-the-art methods using various evaluation protocols.