Social-MAE: Social Masked Autoencoder for Multi-person Motion Representation Learning

• URL: http://arxiv.org/abs/2404.05578v1
• Date: Mon, 8 Apr 2024 14:54:54 GMT
• Title: Social-MAE: Social Masked Autoencoder for Multi-person Motion Representation Learning
• Abstract summary: We introduce Social-MAE, a transformer-based masked autoencoder framework for multi-person human motion data. The framework uses masked modeling to pre-train the encoder to reconstruct masked human joint trajectories. It achieves the state-of-the-art results on various high-level social tasks, including multi-person pose forecasting, social grouping, and social action understanding.
• Score: 16.85316898267404
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: For a complete comprehension of multi-person scenes, it is essential to go beyond basic tasks like detection and tracking. Higher-level tasks, such as understanding the interactions and social activities among individuals, are also crucial. Progress towards models that can fully understand scenes involving multiple people is hindered by a lack of sufficient annotated data for such high-level tasks. To address this challenge, we introduce Social-MAE, a simple yet effective transformer-based masked autoencoder framework for multi-person human motion data. The framework uses masked modeling to pre-train the encoder to reconstruct masked human joint trajectories, enabling it to learn generalizable and data efficient representations of motion in human crowded scenes. Social-MAE comprises a transformer as the MAE encoder and a lighter-weight transformer as the MAE decoder which operates on multi-person joints' trajectory in the frequency domain. After the reconstruction task, the MAE decoder is replaced with a task-specific decoder and the model is fine-tuned end-to-end for a variety of high-level social tasks. Our proposed model combined with our pre-training approach achieves the state-of-the-art results on various high-level social tasks, including multi-person pose forecasting, social grouping, and social action understanding. These improvements are demonstrated across four popular multi-person datasets encompassing both human 2D and 3D body pose.

Related papers

• ACME: A Multi-Cultural, Multi-Embodiment Social-Navigation Dataset [51.60529216343344]
  ACME: A Cross-cultural, Multi-Embodiment dataset for social navigation.<n>ACME is a large-scale data collection effort across 8 sites in 5 countries, using 7 robot embodiments.<n>Unlike prior datasets, it focuses on capturing goal-driven social navigation behavior in complex social scenarios with explicit robot-crowd interaction through robot speech.
  arXiv  Detail & Related papers  (2026-07-24T04:23:21Z)
• MOCHI: Motion Enhancement of Collaborative Human-object Interactions [9.360852329235477]
  Collaborative human-object interaction shows and complex movements that require mutual anticipation and continuous adjustment between participants and the shared object.<n>We present MOCHI, a two-stage framework for enhancing noisy MHOI data.<n>Our approach first generates physically plausible hand grasps through optimization from noisy body input, producing grasps that are both physically plausible and semantically consistent with the body pose.<n>During the optimization process, we introduce optimization objectives to encode human-object and human-human interaction information within these single-person priors.
  arXiv  Detail & Related papers  (2026-06-16T17:58:44Z)
• MuPPet: Multi-person 2D-to-3D Pose Lifting [11.539750714535773]
  We propose MuPPet, a novel multi-person 2D-to-3D pose lifting framework.<n>MuPPet significantly outperforms state-of-the-art single- and multi-person 2D-to-3D pose lifting methods.<n>Our findings highlight the importance of modeling inter-person correlations, paving the way for accurate and socially-aware 3D pose estimation.
  arXiv  Detail & Related papers  (2026-04-08T12:29:33Z)
• Rendering Multi-Human and Multi-Object with 3D Gaussian Splatting [47.35917581534084]
  Reconstructing dynamic scenes with multiple interacting humans and objects from sparse-view inputs is critical for creating high-fidelity digital twins for robotics and VR/AR.<n>This problem, which we term Multi-Human Multi-Object rendering, presents two significant obstacles.<n>We propose MM-GS, a novel hierarchical framework built upon 3D Splatting.
  arXiv  Detail & Related papers  (2026-04-03T12:20:36Z)
• Diffusion Forcing for Multi-Agent Interaction Sequence Modeling [52.769202433667125]
  MAGNet is a unified autoregressive diffusion framework for multi-agent motion generation.<n>It supports a wide range of interaction tasks through flexible conditioning and sampling.<n>It captures both tightly synchronized activities and loosely structured social interactions.
  arXiv  Detail & Related papers  (2025-12-19T18:59:02Z)
• Combo-Gait: Unified Transformer Framework for Multi-Modal Gait Recognition and Attribute Analysis [35.9240903956677]
  We propose a framework that combines 2D temporal silhouettes with 3D SMPL features for robust gait analysis.<n>Beyond identification, we introduce a multitask learning strategy that jointly performs gait recognition and human attribute estimation.<n>Our approach outperforms state-of-the-art methods in gait recognition and provides accurate human attribute estimation.
  arXiv  Detail & Related papers  (2025-10-12T02:56:40Z)
• Social-MAE: A Transformer-Based Multimodal Autoencoder for Face and Voice [8.44136409474895]
  We present Social-MAE, our pre-trained audiovisual Masked Autoencoder based on an extended version of Contrastive Audio-Visual Masked Auto-Encoder (CAV-MAE)<n>We modify CAV-MAE to receive a larger number of frames as input and pre-train it on a large dataset of human social interaction (VoxCeleb2) in a self-supervised manner.<n>The model achieves state-of-the-art results on multimodal emotion recognition and laughter recognition and competitive results for apparent personality estimation.
  arXiv  Detail & Related papers  (2025-08-24T19:49:48Z)
• MAMMA: Markerless & Automatic Multi-Person Motion Action Capture [37.06717786024836]
  MAMMA is a markerless motion-capture pipeline that recovers SMPL-X parameters from multi-view video of two-person interaction sequences.<n>We introduce a method that predicts dense 2D surface landmarks conditioned on segmentation masks.<n>We demonstrate that our approach can handle complex person--person interaction and offers greater accuracy than existing methods.
  arXiv  Detail & Related papers  (2025-06-16T02:04:51Z)
• HumanVBench: Exploring Human-Centric Video Understanding Capabilities of MLLMs with Synthetic Benchmark Data [55.739633494946204]
  We present HumanVBench, an innovative benchmark meticulously crafted to bridge gaps in the evaluation of video MLLMs. HumanVBench comprises 16 carefully designed tasks that explore two primary dimensions: inner emotion and outer manifestations, spanning static and dynamic, basic and complex, as well as single-modal and cross-modal aspects. A comprehensive evaluation across 22 SOTA video MLLMs reveals notable limitations in current performance, especially in cross-modal and emotion perception.
  arXiv  Detail & Related papers  (2024-12-23T13:45:56Z)
• CoMA: Compositional Human Motion Generation with Multi-modal Agents [22.151443524452876]
  CoMA is an agent-based solution for complex human motion generation, editing, and comprehension. Our framework enables generation of both short and long motion sequences with detailed instructions, text-guided motion editing, and self-correction.
  arXiv  Detail & Related papers  (2024-12-10T09:08:41Z)
• CooHOI: Learning Cooperative Human-Object Interaction with Manipulated Object Dynamics [44.30880626337739]
  CooHOI is a framework designed to tackle the challenge of multi-humanoid object transportation problem. A single humanoid character learns to interact with objects through imitation learning from human motion priors. Then, the humanoid learns to collaborate with others by considering the shared dynamics of the manipulated object.
  arXiv  Detail & Related papers  (2024-06-20T17:59:22Z)
• Towards a Unified Transformer-based Framework for Scene Graph Generation and Human-object Interaction Detection [116.21529970404653]
  We introduce SG2HOI+, a unified one-step model based on the Transformer architecture. Our approach employs two interactive hierarchical Transformers to seamlessly unify the tasks of SGG and HOI detection. Our approach achieves competitive performance when compared to state-of-the-art HOI methods.
  arXiv  Detail & Related papers  (2023-11-03T07:25:57Z)
• Human-oriented Representation Learning for Robotic Manipulation [64.59499047836637]
  Humans inherently possess generalizable visual representations that empower them to efficiently explore and interact with the environments in manipulation tasks. We formalize this idea through the lens of human-oriented multi-task fine-tuning on top of pre-trained visual encoders. Our Task Fusion Decoder consistently improves the representation of three state-of-the-art visual encoders for downstream manipulation policy-learning.
  arXiv  Detail & Related papers  (2023-10-04T17:59:38Z)
• Transferring Foundation Models for Generalizable Robotic Manipulation [82.12754319808197]
  We propose a novel paradigm that effectively leverages language-reasoning segmentation mask generated by internet-scale foundation models. Our approach can effectively and robustly perceive object pose and enable sample-efficient generalization learning. Demos can be found in our submitted video, and more comprehensive ones can be found in link1 or link2.
  arXiv  Detail & Related papers  (2023-06-09T07:22:12Z)
• Multimodal Vision Transformers with Forced Attention for Behavior Analysis [0.0]
  We introduce the Forced Attention (FAt) Transformer which utilize forced attention with a modified backbone for input encoding and a use of additional inputs. FAt Transformers are applied to two downstream tasks: personality recognition and body language recognition. We achieve state-of-the-art results for Udiva v0.5, First Impressions v2 and MPII Group Interaction datasets.
  arXiv  Detail & Related papers  (2022-12-07T21:56:50Z)
• MetaGraspNet: A Large-Scale Benchmark Dataset for Vision-driven Robotic Grasping via Physics-based Metaverse Synthesis [78.26022688167133]
  We present a large-scale benchmark dataset for vision-driven robotic grasping via physics-based metaverse synthesis. The proposed dataset contains 100,000 images and 25 different object types. We also propose a new layout-weighted performance metric alongside the dataset for evaluating object detection and segmentation performance.
  arXiv  Detail & Related papers  (2021-12-29T17:23:24Z)
• Multi-Person 3D Motion Prediction with Multi-Range Transformers [16.62864429495888]
  We introduce a Multi-Range Transformers model which contains of a local-range encoder for individual motion and a global-range encoder for social interactions. Our model not only outperforms state-of-the-art methods on long-term 3D motion prediction, but also generates diverse social interactions.
  arXiv  Detail & Related papers  (2021-11-23T18:41:13Z)
• Estimating 3D Motion and Forces of Human-Object Interactions from Internet Videos [49.52070710518688]
  We introduce a method to reconstruct the 3D motion of a person interacting with an object from a single RGB video. Our method estimates the 3D poses of the person together with the object pose, the contact positions and the contact forces on the human body.
  arXiv  Detail & Related papers  (2021-11-02T13:40:18Z)
• Task-Generic Hierarchical Human Motion Prior using VAEs [44.356707509079044]
  A deep generative model that describes human motions can benefit a wide range of fundamental computer vision and graphics tasks. We present a method for learning complex human motions independent of specific tasks using a combined global and local latent space. We demonstrate the effectiveness of our hierarchical motion variational autoencoder in a variety of tasks including video-based human pose estimation.
  arXiv  Detail & Related papers  (2021-06-07T23:11:42Z)
• The IKEA ASM Dataset: Understanding People Assembling Furniture through Actions, Objects and Pose [108.21037046507483]
  IKEA ASM is a three million frame, multi-view, furniture assembly video dataset that includes depth, atomic actions, object segmentation, and human pose. We benchmark prominent methods for video action recognition, object segmentation and human pose estimation tasks on this challenging dataset. The dataset enables the development of holistic methods, which integrate multi-modal and multi-view data to better perform on these tasks.
  arXiv  Detail & Related papers  (2020-07-01T11:34:46Z)

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.