Self-supervised Video Representation Learning by Uncovering Spatio-temporal Statistics

• URL: http://arxiv.org/abs/2008.13426v2
• Date: Fri, 29 Jan 2021 02:41:22 GMT
• Title: Self-supervised Video Representation Learning by Uncovering Spatio-temporal Statistics
• Authors: Jiangliu Wang, Jianbo Jiao, Linchao Bao, Shengfeng He, Wei Liu, and Yun-hui Liu
• Abstract summary: This paper proposes a novel pretext task to address the self-supervised learning problem. We compute a series of partitioning-temporal statistical summaries, such as the spatial location and dominant direction of the largest motion. A neural network is built and trained to yield the statistical summaries given the video frames as inputs.
• Score: 74.6968179473212
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper proposes a novel pretext task to address the self-supervised video representation learning problem. Specifically, given an unlabeled video clip, we compute a series of spatio-temporal statistical summaries, such as the spatial location and dominant direction of the largest motion, the spatial location and dominant color of the largest color diversity along the temporal axis, etc. Then a neural network is built and trained to yield the statistical summaries given the video frames as inputs. In order to alleviate the learning difficulty, we employ several spatial partitioning patterns to encode rough spatial locations instead of exact spatial Cartesian coordinates. Our approach is inspired by the observation that human visual system is sensitive to rapidly changing contents in the visual field, and only needs impressions about rough spatial locations to understand the visual contents. To validate the effectiveness of the proposed approach, we conduct extensive experiments with four 3D backbone networks, i.e., C3D, 3D-ResNet, R(2+1)D and S3D-G. The results show that our approach outperforms the existing approaches across these backbone networks on four downstream video analysis tasks including action recognition, video retrieval, dynamic scene recognition, and action similarity labeling. The source code is publicly available at: https://github.com/laura-wang/video_repres_sts.

Related papers

• ALSTER: A Local Spatio-Temporal Expert for Online 3D Semantic Reconstruction [62.599588577671796]
  We propose an online 3D semantic segmentation method that incrementally reconstructs a 3D semantic map from a stream of RGB-D frames. Unlike offline methods, ours is directly applicable to scenarios with real-time constraints, such as robotics or mixed reality.
  arXiv  Detail & Related papers  (2023-11-29T20:30:18Z)
• Semi-supervised 3D Video Information Retrieval with Deep Neural Network and Bi-directional Dynamic-time Warping Algorithm [14.39527406033429]
  The proposed algorithm is designed to handle large video datasets and retrieve the most related videos to a given inquiry video clip. We split both the candidate and the inquiry videos into a sequence of clips and convert each clip to a representation vector using an autoencoder-backed deep neural network. We then calculate a similarity measure between the sequences of embedding vectors using a bi-directional dynamic time-warping method.
  arXiv  Detail & Related papers  (2023-09-03T03:10:18Z)
• Neural Implicit Dense Semantic SLAM [83.04331351572277]
  We propose a novel RGBD vSLAM algorithm that learns a memory-efficient, dense 3D geometry, and semantic segmentation of an indoor scene in an online manner. Our pipeline combines classical 3D vision-based tracking and loop closing with neural fields-based mapping. Our proposed algorithm can greatly enhance scene perception and assist with a range of robot control problems.
  arXiv  Detail & Related papers  (2023-04-27T23:03:52Z)
• Video-Text Pre-training with Learned Regions [59.30893505895156]
  Video-Text pre-training aims at learning transferable representations from large-scale video-text pairs. We propose a module for videotext-learning, RegionLearner, which can take into account the structure of objects during pre-training on large-scale video-text pairs.
  arXiv  Detail & Related papers  (2021-12-02T13:06:53Z)
• HighlightMe: Detecting Highlights from Human-Centric Videos [52.84233165201391]
  We present a domain- and user-preference-agnostic approach to detect highlightable excerpts from human-centric videos. We use an autoencoder network equipped with spatial-temporal graph convolutions to detect human activities and interactions. We observe a 4-12% improvement in the mean average precision of matching the human-annotated highlights over state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-10-05T01:18:15Z)
• Spatial-Temporal Correlation and Topology Learning for Person Re-Identification in Videos [78.45050529204701]
  We propose a novel framework to pursue discriminative and robust representation by modeling cross-scale spatial-temporal correlation. CTL utilizes a CNN backbone and a key-points estimator to extract semantic local features from human body. It explores a context-reinforced topology to construct multi-scale graphs by considering both global contextual information and physical connections of human body.
  arXiv  Detail & Related papers  (2021-04-15T14:32:12Z)
• LIGHTEN: Learning Interactions with Graph and Hierarchical TEmporal Networks for HOI in videos [13.25502885135043]
  Analyzing the interactions between humans and objects from a video includes identification of relationships between humans and the objects present in the video. We present a hierarchical approach, LIGHTEN, to learn visual features to effectively capture truth at multiple granularities in a video. We achieve state-of-the-art results in human-object interaction detection (88.9% and 92.6%) and anticipation tasks of CAD-120 and competitive results on image based HOI detection in V-COCO.
  arXiv  Detail & Related papers  (2020-12-17T05:44:07Z)
• Visual Descriptor Learning from Monocular Video [25.082587246288995]
  We propose a novel way to estimate dense correspondence on an RGB image by training a fully convolutional network. Our method learns from RGB videos using contrastive loss, where relative labeling is estimated from optical flow. Not only does the method perform well on test data with the same background, it also generalizes to situations with a new background.
  arXiv  Detail & Related papers  (2020-04-15T11:19:57Z)

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.