Monocular depth estimation is scale-ambiguous, and thus requires scale
s...
3D object detection from visual sensors is a cornerstone capability of
r...
This work proposes an end-to-end multi-camera 3D multi-object tracking (...
A key contributor to recent progress in 3D detection from single images ...
Optical sensors and learning algorithms for autonomous vehicles have
dra...
Test-time adaptation is a special setting of unsupervised domain adaptat...
Multi-frame depth estimation improves over single-frame approaches by al...
In this paper, we present a system to train driving policies from experi...
Active learning for object detection is conventionally achieved by apply...
Transfer learning eases the burden of training a well-performed model fr...
We learn an interactive vision-based driving policy from pre-recorded dr...
Vision-based urban driving is hard. The autonomous system needs to learn...
We present an approach for building an active agent that learns to segme...
The current dominant paradigm for imitation learning relies on strong
su...
Manipulation of deformable objects, such as ropes and cloth, is an impor...