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

2026-08-21

AI Summary

为了解决现有模型无法处理动态感知和预测的问题，本文提出了Human-JEPA模型，通过锚定预测方法在视频上进行训练，从而实现对人类行为的理解。

Abstract

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.

Problem

• human-centric vision model
• perception
• anticipation
• motion

Innovation

• Human-JEPA
• anchored forecasting
• dense perception
• past-to-future split
• anticipation