Human drivers can seamlessly adapt their driving decisions across geographical locations
with diverse conditions and rules of the road, eg, left vs. right-hand traffic. In contrast, existing …

In the era of the COVID-19 pandemic, the surge in information sharing on social media,
particularly Twitter, necessitates a nuanced understanding of sentiments. Conventional …

Abstract We introduce Motion Diversification Networks a novel framework for learning to
generate realistic and diverse 3D human motion. Despite recent advances in deep …

Estimating relative camera poses between images has been a central problem in computer
vision. Methods that find correspondences and solve for the fundamental matrix offer high …

While behavior cloning has recently emerged as a highly successful paradigm for
autonomous driving humans rarely learn to perform complex tasks such as driving via …

We present a highly scalable self-training framework for incrementally adapting vision-
based end-to-end autonomous driving policies in a semi-supervised manner ie over a …

Data-driven visual-inertial odometry (VIO) has received highlights for its performance since
VIOs are a crucial compartment in autonomous robots. However, their deployment on …

World models can foresee the outcomes of different actions, which is of paramount
importance for autonomous driving. Nevertheless, existing driving world models still have …

Embodied vision-based real-world systems, such as mobile robots, require a careful
balance between energy consumption, compute latency, and safety constraints to optimize …

Effectively navigating a dynamic 3D world requires a comprehensive understanding of the
3D geometry and motion of surrounding objects and layouts. However, existing methods for …