Current state-of-the-art point cloud-based perception methods usually re...
Point-, voxel-, and range-views are three representative forms of point
...
Human-centric scene understanding is significant for real-world applicat...
Zero-shot point cloud segmentation aims to make deep models capable of
r...
Vision foundation models such as Contrastive Vision-Language Pre-trainin...
In this paper, we propose a novel self-supervised motion estimator for
L...
Current on-board chips usually have different computing power, which mea...
Motion capture is a long-standing research problem. Although it has been...
We present SLOPER4D, a novel scene-aware dataset collected in large urba...
Training deep models for semantic scene completion (SSC) is challenging ...
LiDAR segmentation is crucial for autonomous driving perception. Recent
...
Contrastive language-image pre-training (CLIP) achieves promising result...
Domain adaptation for Cross-LiDAR 3D detection is challenging due to the...
Depth estimation is usually ill-posed and ambiguous for monocular
camera...
LiDAR can capture accurate depth information in large-scale scenarios wi...
HD map reconstruction is crucial for autonomous driving. LiDAR-based met...
We investigate transductive zero-shot point cloud semantic segmentation ...
Predicting the future motion of road participants is crucial for autonom...
Vision-centric BEV perception has recently received increased attention ...
This article addresses the problem of distilling knowledge from a large
...
We propose a multi-sensor fusion method for capturing challenging 3D hum...
Accurately detecting and tracking pedestrians in 3D space is challenging...
Existing motion capture datasets are largely short-range and cannot yet ...
Real scans always miss partial geometries of objects due to the
self-occ...
Promising performance has been achieved for visual perception on the poi...
We propose Human-centered 4D Scene Capture (HSC4D) to accurately and
eff...
Emerging Metaverse applications demand reliable, accurate, and photoreal...
Recently, records on stereo matching benchmarks are constantly broken by...
State-of-the-art methods for driving-scene LiDAR-based perception (inclu...
3D detection plays an indispensable role in environment perception. Due ...
Detecting 3D landmarks on cone-beam computed tomography (CBCT) is crucia...
Well-annotated medical images are costly and sometimes even impossible t...
Markerless motion capture and understanding of professional non-daily hu...
A thorough and holistic scene understanding is crucial for autonomous
ve...
Capturing challenging human motions is critical for numerous application...
State-of-the-art methods for large-scale driving-scene LiDAR segmentatio...
State-of-the-art methods for large-scale driving-scene LiDAR semantic
se...
Current methods for trajectory prediction operate in supervised manners,...
Multi-class 3D object detection aims to localize and classify objects of...
This paper reviews the CVPR 2019 challenge on Autonomous Driving. Baidu'...
High-level (e.g., semantic) features encoded in the latter layers of
con...
Marking anatomical landmarks in cephalometric radiography is a critical
...
Simulation systems have become an essential component in the development...
To safely and efficiently navigate in complex urban traffic, autonomous
...
We present a novel algorithm for computing collision-free navigation for...
We present a novel algorithm for reciprocal collision avoidance between
...