mlsanityData to 14 Oct 2026

Human-JEPA: A Human-Centric Vision Model that Perceives and Anticipates

arXiv 2608.21160 ↗First posted 21 Aug 2026cs.CV cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Machines that understand humans should perceive the present and anticipate the future. Existing human-centric vision model are pretrained on human images, set the state of the art in static dense perception, so motion and anticipation are out of reach. Here we present Human-JEPA, a human-centric vision model trained on video by anchored forecasting: dense targets are pinned to a frozen copy of the initialization, preventing a silent collapse of dense perception, and block masks are replaced by a pure past-to-future split, avoiding a five-point action tax and a seventeen-point re-identification collapse. Under frozen probes, Human-JEPA leads the pixel-anchored specialists on pose and person re-identification at 2.7 times fewer parameters, conceding high-resolution dense parsing, and its released predictor head is the first that does not degrade anticipation. A single safely adapted model thus serves both halves of understanding humans.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 42 of its 51 references that are in the map. First posted on arXiv 21 Aug 2026.

Builds on

42 of 51 references.

References · 42

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  4. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  5. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  6. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  7. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  8. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  9. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  10. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,287
  11. 2021Vision Transformers for Dense Predictioncited by 1,803
  12. 2021An Empirical Study of Training Self-Supervised Vision Transformerscited by 1,616
30 more
  1. 2025DINOv3cited by 1,414
  2. 2019Contrastive Multiview Codingcited by 1,301
  3. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 1,039
  4. 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningcited by 809
  5. 2022data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Languagecited by 727
  6. 2018DensePose: Dense Human Pose Estimation In The Wildcited by 575
  7. 2019NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity Understandingcited by 533
  8. 2021AI Choreographer: Music Conditioned 3D Dance Generation with AIST++cited by 485
  9. 2019A Short Note on the Kinetics-700 Human Action Datasetcited by 310
  10. 2026LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixelscited by 273
  11. 2025LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristicscited by 254
  12. 2024Sapiens: Foundation for Human Vision Modelscited by 229
  13. 2020Deep Learning-Based Human Pose Estimation: A Surveycited by 207
  14. 2023A Survey on Self-supervised Learning: Algorithms, Applications, and Future Trendscited by 207
  15. 2026V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learningcited by 117
  16. 2022Efficient Self-supervised Learning with Contextualized Target Representations for Vision, Speech and Languagecited by 110
  17. 2023UniHCP: A Unified Model for Human-Centric Perceptionscited by 59
  18. 2023MC-JEPA: A Joint-Embedding Predictive Architecture for Self-Supervised Learning of Motion and Content Featurescited by 57
  19. 2015Deep Human Parsing with Active Template Regressioncited by 51
  20. 2023HumanBench: Towards General Human-centric Perception with Projector Assisted Pretrainingcited by 38
  21. 2023HAP: Structure-Aware Masked Image Modeling for Human-Centric Perceptioncited by 33
  22. 2024Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloudcited by 30
  23. 2023Hulk: A Universal Knowledge Translator for Human-Centric Taskscited by 22
  24. 2023Unified Pre-training with Pseudo Texts for Text-To-Image Person Re-identificationcited by 19
  25. 2025Audio-JEPA: Joint-Embedding Predictive Architecture for Audio Representation Learningcited by 17
  26. 20243D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learningcited by 14
  27. 2025Self-Supervised Representation Learning with Joint Embedding Predictive Architecture for Automotive LiDAR Object Detectioncited by 13
  28. 2026ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Modelcited by 13
  29. 2025Human-Centric Foundation Models: Perception, Generation and Agentic Modelingcited by 12
  30. 2023Deep Learning for Human Parsing: A Surveycited by 1

Built on by

Not cited yet.