Human‑JEPA is a human‑centric vision model trained on video that uses anchored forecasting to prevent silent collapse of dense perception. By replacing block masks with a pure past‑to‑future split, it avoids action‑tax and re‑identification collapse, enabling strong performance on pose and person re‑identification with far fewer parameters. The released predictor head uniquely preserves anticipation capability, allowing a single model to handle both perception and anticipation of humans.
arXiv:2608.20974v1 Announce Type: cross Abstract: Video Joint Embedding Predictive Architecture (V-JEPA) learns powerful spatiotemporal representations from video through self-supervised latent featu...
arXiv:2607. 09024v1 Announce Type: cross Abstract: Driven by next-token prediction, NLP shifted from task-specific models into powerful generalist foundation models.
arXiv:2602. 14771v5 Announce Type: replace-cross Abstract: The human visual system tracks objects by integrating current observations with previously observed information, adapting to target and scene changes, and reasoning about occlusion at fine granularity.
LeVJEPA is a video encoder that eliminates the need for architectural asymmetries, exponential-moving-average target encoders, stop-gradients, and capacity-limited predictors used in prior self‑supervised methods. It trains a single encoder with an invariance loss over global and local views, regularized by SIGReg to prevent collapse, and achieves strong performance with far less pretraining compute. The approach also allows block‑causal attention, making temporal ordering a property of the encoder itself, and matches or surpasses state‑of‑the‑art baselines on both appearance‑centric and motion‑centric benchmarks.
MC-DeTra is a reimplementation of the DeTra model that jointly performs object detection and socially-aware trajectory forecasting in bird's-eye-view images. It introduces motion-consistency mechanisms that add supervision from each actor’s past motion, surrounding traffic occupancy, and a consistency constraint aligning predicted heading with motion direction. The added losses are train‑only and inference‑safe, improving dynamic trajectory forecasting on the Waymo Open Dataset while maintaining or enhancing detection accuracy.
arXiv:2609.21379v1 Announce Type: new Abstract: Accurate traffic forecasting requires both understanding scene dynamics and synthesizing realistic future observations. Recent diffusion-based video ge...
The paper introduces a redundancy-aware fusion framework for EgoExo proficiency estimation, which integrates fine-grained motion cues from egocentric views with spatial context from exocentric views. It identifies multiview redundancy and overfitting as key challenges and proposes two modules—AdaMVS for adaptive view selection and VIB-GB for compressing redundant signals—to address them. Experiments on EgoExo-4D and EgoExo-Fitness show that the method learns to select informative views and fuse them effectively, achieving state‑of‑the‑art results.
arXiv:2607. 08436v1 Announce Type: cross Abstract: Egocentric human data offers scalable supervision for robot manipulation.
World in World introduces a training‑free, inference‑time interface that transforms diverse control signals—such as source‑video observations, target‑view projections, geometry renderings, and retrieved states—into camera‑ and time‑labelled visual states. These states are processed by a frozen causal video model’s self‑attention, enabling tasks like camera‑controlled rerendering, long‑horizon revisiting, and human‑motion transfer without additional training. The method employs a correspondence router and evidence‑wise attention to align token identities and regulate auxiliary channel contributions during a single denoising pass.
arXiv:2607. 02404v1 Announce Type: cross Abstract: Image encoders trained with LeJEPA can deliver strong features for downstream tasks, but, like other image-level self-supervised methods, typically require large training datasets.
arXiv:2605.15618v2 Announce Type: replace-cross Abstract: Self-supervised video models are increasingly framed as world models, yet they are still evaluated almost entirely on clean video and reporte...
arXiv:2606. 06885v1 Announce Type: cross Abstract: Human Image Animation has seen significant advancements, primarily driven by diffusion models.