Skeleton2vec: A Self-supervised Learning Framework with Contextualized Target Representations for Skeleton Sequence

• URL: http://arxiv.org/abs/2401.00921v1
• Date: Mon, 1 Jan 2024 12:08:35 GMT
• Title: Skeleton2vec: A Self-supervised Learning Framework with Contextualized Target Representations for Skeleton Sequence
• Authors: Ruizhuo Xu, Linzhi Huang, Mei Wang, Jiani Hu, Weihong Deng
• Abstract summary: We show that using high-level contextualized features as prediction targets can achieve superior performance. Specifically, we propose Skeleton2vec, a simple and efficient self-supervised 3D action representation learning framework. Our proposed Skeleton2vec outperforms previous methods and achieves state-of-the-art results.
• Score: 56.092059713922744
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Self-supervised pre-training paradigms have been extensively explored in the field of skeleton-based action recognition. In particular, methods based on masked prediction have pushed the performance of pre-training to a new height. However, these methods take low-level features, such as raw joint coordinates or temporal motion, as prediction targets for the masked regions, which is suboptimal. In this paper, we show that using high-level contextualized features as prediction targets can achieve superior performance. Specifically, we propose Skeleton2vec, a simple and efficient self-supervised 3D action representation learning framework, which utilizes a transformer-based teacher encoder taking unmasked training samples as input to create latent contextualized representations as prediction targets. Benefiting from the self-attention mechanism, the latent representations generated by the teacher encoder can incorporate the global context of the entire training samples, leading to a richer training task. Additionally, considering the high temporal correlations in skeleton sequences, we propose a motion-aware tube masking strategy which divides the skeleton sequence into several tubes and performs persistent masking within each tube based on motion priors, thus forcing the model to build long-range spatio-temporal connections and focus on action-semantic richer regions. Extensive experiments on NTU-60, NTU-120, and PKU-MMD datasets demonstrate that our proposed Skeleton2vec outperforms previous methods and achieves state-of-the-art results.

Related papers

• Multi-Level Contrastive Learning for Dense Prediction Task [59.591755258395594]
  We present Multi-Level Contrastive Learning for Dense Prediction Task (MCL), an efficient self-supervised method for learning region-level feature representation for dense prediction tasks. Our method is motivated by the three key factors in detection: localization, scale consistency and recognition. Our method consistently outperforms the recent state-of-the-art methods on various datasets with significant margins.
  arXiv  Detail & Related papers  (2023-04-04T17:59:04Z)
• Self-Supervised Representation Learning from Temporal Ordering of Automated Driving Sequences [49.91741677556553]
  We propose TempO, a temporal ordering pretext task for pre-training region-level feature representations for perception tasks. We embed each frame by an unordered set of proposal feature vectors, a representation that is natural for object detection or tracking systems. Extensive evaluations on the BDD100K, nuImages, and MOT17 datasets show that our TempO pre-training approach outperforms single-frame self-supervised learning methods.
  arXiv  Detail & Related papers  (2023-02-17T18:18:27Z)
• Self-supervised Action Representation Learning from Partial Spatio-Temporal Skeleton Sequences [29.376328807860993]
  We propose a Partial Spatio-Temporal Learning (PSTL) framework to exploit the local relationship between different skeleton joints and video frames. Our method achieves state-of-the-art performance on NTURGB+D 60, NTURGBMM+D 120 and PKU-D under various downstream tasks.
  arXiv  Detail & Related papers  (2023-02-17T17:35:05Z)
• Hierarchically Self-Supervised Transformer for Human Skeleton Representation Learning [45.13060970066485]
  We propose a self-supervised hierarchical pre-training scheme incorporated into a hierarchical Transformer-based skeleton sequence encoder (Hi-TRS) Under both supervised and semi-supervised evaluation protocols, our method achieves the state-of-the-art performance.
  arXiv  Detail & Related papers  (2022-07-20T04:21:05Z)
• Learning from Temporal Spatial Cubism for Cross-Dataset Skeleton-based Action Recognition [88.34182299496074]
  Action labels are only available on a source dataset, but unavailable on a target dataset in the training stage. We utilize a self-supervision scheme to reduce the domain shift between two skeleton-based action datasets. By segmenting and permuting temporal segments or human body parts, we design two self-supervised learning classification tasks.
  arXiv  Detail & Related papers  (2022-07-17T07:05:39Z)
• Joint-bone Fusion Graph Convolutional Network for Semi-supervised Skeleton Action Recognition [65.78703941973183]
  We propose a novel correlation-driven joint-bone fusion graph convolutional network (CD-JBF-GCN) as an encoder and use a pose prediction head as a decoder. Specifically, the CD-JBF-GC can explore the motion transmission between the joint stream and the bone stream. The pose prediction based auto-encoder in the self-supervised training stage allows the network to learn motion representation from unlabeled data.
  arXiv  Detail & Related papers  (2022-02-08T16:03:15Z)
• UNIK: A Unified Framework for Real-world Skeleton-based Action Recognition [11.81043814295441]
  We introduce UNIK, a novel skeleton-based action recognition method that is able to generalize across datasets. To study the cross-domain generalizability of action recognition in real-world videos, we re-evaluate state-of-the-art approaches as well as the proposed UNIK. Results show that the proposed UNIK, with pre-training on Posetics, generalizes well and outperforms state-of-the-art when transferred onto four target action classification datasets.
  arXiv  Detail & Related papers  (2021-07-19T02:00:28Z)
• Semi-supervised Facial Action Unit Intensity Estimation with Contrastive Learning [54.90704746573636]
  Our method does not require to manually select key frames, and produces state-of-the-art results with as little as $2%$ of annotated frames. We experimentally validate that our method outperforms existing methods when working with as little as $2%$ of randomly chosen data.
  arXiv  Detail & Related papers  (2020-11-03T17:35:57Z)
• Action Localization through Continual Predictive Learning [14.582013761620738]
  We present a new approach based on continual learning that uses feature-level predictions for self-supervision. We use a stack of LSTMs coupled with CNN encoder, along with novel attention mechanisms, to model the events in the video and use this model to predict high-level features for the future frames. This self-supervised framework is not complicated as other approaches but is very effective in learning robust visual representations for both labeling and localization.
  arXiv  Detail & Related papers  (2020-03-26T23:32:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.