End-to-end autonomous driving aims to build a fully differentiable syste...
Registration of distant outdoor LiDAR point clouds is crucial to extendi...
The autonomous driving community has witnessed a rapid growth in approac...
Human driver can easily describe the complex traffic scene by visual sys...
Recently, video object segmentation (VOS) referred by multi-modal signal...
End-to-end autonomous driving has made impressive progress in recent yea...
With the prevalence of multimodal learning, camera-LiDAR fusion has gain...
Multi-camera 3D object detection for autonomous driving is a challenging...
Driving scenes are extremely diverse and complicated that it is impossib...
Masked Autoencoders (MAE) have been popular paradigms for large-scale vi...
Witnessing the impressive achievements of pre-training techniques on
lar...
Modern autonomous driving system is characterized as modular tasks in
se...
We present a novel bird's-eye-view (BEV) detector with perspective
super...
Masked Autoencoders (MAE) have been prevailing paradigms for large-scale...
Establishment of point correspondence between camera and object coordina...
Many existing autonomous driving paradigms involve a multi-stage discret...
Current end-to-end autonomous driving methods either run a controller ba...
One essential task for autonomous driving is to encode the information o...
3D visual perception tasks, including 3D detection and map segmentation ...
As a basic component of SE(3)-equivariant deep feature learning, steerab...
Monocular 3D object detection is an important task in autonomous driving...
Most differentiable neural architecture search methods construct a super...
We present a new learning-based approach to recover egocentric 3D vehicl...
Neural architecture search (NAS) aims to produce the optimal sparse solu...
A recent approach for object detection and human pose estimation is to
r...
The correspondence between residual networks and dynamical systems motiv...
The panoptic segmentation task requires a unified result from semantic a...
Few-shot learning is an important area of research. Conceptually, humans...
A well-trained model should classify objects with a unanimous score for ...
Gliomas are the most common primary brain malignancies, with different
d...
A capsule is a collection of neurons which represents different variants...
Feature matters. How to train a deep network to acquire discriminative
f...
In this paper, we propose a zoom-out-and-in network for generating objec...
Since convolutional neural network (CNN) lacks an inherent mechanism to
...
Person recognition aims at recognizing the same identity across time and...
Visual tracking addresses the problem of identifying and localizing an
u...
As a widely used non-linear activation, Rectified Linear Unit (ReLU)
sep...
In this paper, we propose a novel label propagation based method for sal...