Computer Vision

Abstract: The field of 4D world modeling - aiming to jointly capture spatial geometry and temporal dynamics - has witnessed remarkable progress in recent years, driven by advances in large-scale generative models and multimodal learning.

Abstract: We present WinT3R, a feed-forward reconstruction model capable of online prediction of precise camera poses and high-quality point maps. Previous methods suffer from a trade-off between reconstruction quality and real-time performance.

Abstract: We introduce ${\pi }^3$, a feed-forward neural network that offers a novel approach to visual geometry reconstruction, breaking the reliance on a conventional fixed reference view. Previous methods often anchor their reconstructions to a designated viewpoint, an inductive bias that can lead to instability and failures if the reference is suboptimal.

Abstract: In this paper, we introduce an innovative vector quantization based action tokenizer built upon the largest-scale action trajectory dataset to date, leveraging over 100 times more data than previous approaches.

Abstract: World models serve as essential building blocks toward Artificial General Intelligence (AGI), enabling intelligent agents to predict future states and plan actions by simulating complex physical interactions. However, existing interactive models primarily predict visual observations, thereby neglecting crucial hidden states like geometric structures and spatial coherence.

A geometric-aware unified world model, capable of 4D reconstruction, action-conditioned prediction, and visual planning.

A novel representation learning framework that emphasizes the importance of 3D spatial awareness in embodied AI.

A general 3D pre-training approach establishing a pathway to 3D foundational models.

Abstract: In the context of autonomous driving, the significance of effective feature learning is widely acknowledged. While conventional 3D self-supervised pre-training methods have shown widespread success, most methods follow the ideas originally designed for 2D images.

Abstract: Computer vision has emerged as a powerful tool to elevate behavioral research. This protocol describes a computer vision machine learning pipeline called AlphaTracker, which has minimal hardware requirements and produces reliable tracking of multiple unmarked animals, as well as behavioral clustering.