arXiv:2606. 14765v1 Announce Type: cross Abstract: Self-supervised video representation learning has recently advanced through contrastive learning, masked reconstruction, and predictive representation learning.
arXiv:2607. 00514v1 Announce Type: cross Abstract: Automatic understanding of dynamic 4D point clouds, the 3D-point sequences captured over time by depth sensors and LiDAR, is central to robotics and embodied perception.
World models enable agents to perform forward rollout and planning without real-world interaction. However, their application in open-world embodied intelligence remains limited by the high cost of action annotations and the heterogeneity of action spaces across platforms.
arXiv:2609.23881v1 Announce Type: new Abstract: Joint Embedding Predictive Architectures (JEPAs) are a promising paradigm for learning task-agnostic latent world models without visual reconstruction....
arXiv:2608.30279v1 Announce Type: new Abstract: Point cloud video representation learning is crucial for 3D dynamic scene understanding. In this paper, we propose MoSaiC, a novel Motion-Saliency Comp...
GeoLAM is a framework that learns geometry‑grounded latent actions from unlabeled human videos. It uses future‑frame reconstruction with a frozen geometric feature hierarchy and motion supervision from a 4D geometry teacher to capture 3D displacement, image‑plane motion, and surface‑orientation changes. After pretraining, the representation serves as transition targets for a world‑action model trained on robot demonstrations, enabling denoised latent actions and executable action chunks without requiring hand‑pose annotations or future‑video generation during deployment.
arXiv:2606. 07687v1 Announce Type: cross Abstract: Video world models are increasingly used to provide predictive visual representations, yet it remains unclear which pretraining signals induce action-relevant structure in their latent spaces.
arXiv:2603.25165v3 Announce Type: replace Abstract: Recent advances in self-supervised learning (SSL) for point clouds have substantially improved 3D scene understanding without human annotations. Ex...
The paper introduces Skeleton-Language feature Pooling Switching, a weakly‑supervised vision‑language pretraining strategy for skeleton‑based zero‑shot spatio‑temporal action localization. It replaces video‑level pooling with instance‑level feature computation during inference, enabling the model to estimate unseen actions without costly annotations. Additionally, Scene‑Mixed Discriminative Contrastive Learning is proposed to separate actions at the instance level within mixed scenes using a MIL framework, and experiments on four public datasets confirm the method’s effectiveness.
Real-world spatial intelligence requires agents to understand scenes from continuous video streams, where objects move, persist, disappear, and reappear over time. While recent spatial foundation models have enabled generalizable feed-forward 3D reconstruction, most streaming methods remain geometry-centric and lack temporally consistent object-level understanding.
ParticleSplat is a self‑supervised, object‑centric learning framework that extends the Deep Latent Particles (DLP) model into 3D by representing scenes as latent particles mapped to 3D Gaussian splats. It jointly encodes multiple camera views into a shared 3D latent space, enabling unsupervised learning of object masks and controllable 3D scene editing such as moving objects by manipulating latent particles. Experiments on simulated and real‑world datasets demonstrate that this 3D representation improves performance on downstream robotic manipulation tasks.
Multimodal Large Language Models (MLLMs) excel at 2D semantic understanding but lack intrinsic 3D awareness, resulting in representations that fail to maintain geometric and spatial consistency across video frames. Given the scarcity of large-scale 3D data, we present GeoVR, a novel framework that learns geometric representations using purely 2D video sequences.