Autonomous Driving

About the Workshop

The CVPR 2020 Workshop on Autonomous Driving (WAD) aims to gather researchers and engineers from academia and industry to discuss the latest advances in perception for autonomous driving. In this one-day workshop, we will have regular paper presentations, invited speakers, and technical benchmark challenges to present the current state of the art, as well as the limitations and future directions for computer vision in autonomous driving, arguably the most promising application of computer vision and AI in general. The previous chapters of the workshop at CVPR attracted hundreds of researchers to attend. This year, multiple industry sponsors also join our organizing efforts to push its success to a new level.

Program

Our workshop consists of 3 parts arranged in one main program block, the block starts with our two challenges the Argovers and the BDD 100K challenge. Where the challenges are presented and the winning teams present their methods. This is followed by a panel discussion with our invited speakers, where the audience can ask specific questions about their talks but also general questions where the field of autonomous driving is heading. Finally, we have a Q&A session with our paper authors.

This main block is repeated twice, the first block starts at 9 am (PDT) and the block is then repeated at 9 pm (PDT). Note that not all invited speakers will join both sessions. Please also note that not all talks are hosted on the CVPR homepage and that several talks will be hosted at our Workshop homepage http://cvpr2020.wad.vision 

Teaser picture for paper
While a world densely populated with self-driving vehicles (SDVs) might seem futuristic, these vehicles will soon be the norm. They will provide safer
    Authors: Raquel Urtasun   
    Keywords:  Self-driving, V2V, vehicle to vehicle communication, 3D perception, motion forecasting, multi-agent robotics, deep learning
Frid Jun19  
All day
Favorite
Teaser picture for paper
Autonomous Driving
    Authors: Bo Li   
    Keywords:  Autonomous Driving
Frid Jun19  
All day
Favorite
Teaser picture for paper
Learning Robust Driving Policies I will present two recent results on learning robust driving policies that lead to state-of-the-art performance in t
    Authors: Andreas Geiger   
    Keywords:  Autonomous Driving Learning driving polcies Imitation Learning
Frid Jun19  
All day
Favorite
Teaser picture for paper
Computer vision is undergoing a period of rapid progress, rekindling the relationship between perception, action, and cognition. Such connections may
    Authors: Deva Ramanan   
    Keywords:  Autonomous Driving Robotics in the wild Perceptual Understanding
Frid Jun19  
All day
Favorite
Teaser picture for paper
Self-driving cars currently on the road are equipped with dozens of sensors of several types (lidar, radar, sonar, cameras, . . . ). All of this exi
    Authors: Emilio Frazzoli   
    Keywords:  Autonomous Driving Multi-modal perception Wormhole learning
Frid Jun19  
All day
Favorite
Teaser picture for paper
Autonomous Driving
    Authors: Drago Anguelov   
    Keywords:  Autonomous Driving
Frid Jun19  
All day
Favorite
Teaser picture for paper
I will discuss recent achievements on boosting the performance of direct methods for visual simultaneous localization and mapping using deep networks.
    Authors: Daniel Cremers   
    Keywords:  visual SLAM, camera-based reconstruction, direct methods, deep learning, 3D reconstruction, odometry, camera tracking, optimization, structure from motion, scene understanding
Frid Jun19  
All day
Favorite
Teaser picture for paper
In this talk, Kodiak Robotics’ VP of Engineering, Andreas Wendel, gives insights into the development of self-driving semi-trucks. Starting from the
    Authors: Andreas Wendel   
    Keywords:  Self-driving truck, autonomous vehicle, AV, Kodiak, long-range perception, sensing
Frid Jun19  
All day
Favorite
Teaser picture for paper
In this talk, I will speak about two important challenges that must be overcome to turn the current ‘somewhat automated’ into full automation and
    Authors: Dengxin Dai   
    Keywords:  adverse weather/lighting conditions, all-season driving, domain adaptation, social driving
Frid Jun19  
All day
Favorite
Teaser picture for paper
We proposed a monocular 3D object detection method named SMOKE. Our method estimates a 3D bounding box for each object by combining a single detect
    Authors: Zechen Liu; Zizhang Wu; Roland Toth   
    Keywords:  3D Object Detection,Autonomous Driving
Frid Jun19  
All day
Favorite
Teaser picture for paper
With wasserstein loss, the performance of object detection is improved significantly especially for important categories.
    Authors: Xiaofeng Liu; Jane You; Yutao Ren; Risheng Liu; Zhenfei Sheng; Xu Han; Yuzhuo Han; Zhongxuan Luo   
    Keywords:  Object detection, Autonomious driving, Wasserstein training, Appearance similarity
Frid Jun19  
All day
Favorite
Teaser picture for paper
3D object detection from a single image without LiDAR is a challenging task due to the lack of accurate depth information. Conventional 2D convolution
    Authors: Mingyu Ding; Yuqi Huo; Hongwei Yi; Zhe Wang; Jianping Shi; Zhiwu Lu; Ping Luo   
    Keywords:  Monocular, 3D Object Detection, Depth-Guided, Dynamic Local Convolution
Frid Jun19  
All day
Favorite
Teaser picture for paper
In this work, we explore the use of feudal networks to devise a vehicle agent to predict steering angles from first person, dash-cam images of the Uda
    Authors: Faith M Johnson; Kristin Dana   
    Keywords:  Feudal Networks, Autonomous Driving, Hierarchical Learning, Temporal Abstraction
Frid Jun19  
All day
Favorite
Teaser picture for paper
We present a self-supervised learning framework to estimate the individual object motion and monocular depth from video captured by a moving camera.
    Authors: Qi Dai; Vaishakh Patil; Simon Hecker; Dengxin Dai ; Luc Van Gool; Konrad Schindler   
    Keywords:  Motion Estimation;Depth Estimation;Self-supervised Learning;View Synthesis
Frid Jun19  
All day
Favorite
Teaser picture for paper
In autonomous driving, detecting reliable and accurate lane marker positions is a crucial yet challenging task. The conventional approaches for the l
    Authors: Seungwoo Yoo; Heeseok Lee; Heesoo Myeong; Sungrack Yun; Hyoungwoo Park; Janghoon Cho; Duckhoon Kim   
    Keywords:  Lane marker recognition, Autonomous Driving, ADAS, Computer Vision, Machine Learning
Frid Jun19  
All day
Favorite
Teaser picture for paper
This work improves the 3D sensing capabilities of self-driving cars by fusing input from stereo cameras, LiDARs, and single-photon LiDARs.
    Authors: Talha Ahmad Siddiqui; Rishi Madhok; Matthew O'Toole   
    Keywords:  Sensor Fusion, 3D Imaging, Single-Photon LiDAR, Depth Estimation
Frid Jun19  
All day
Favorite
Teaser picture for paper
In this paper, we introduce a two-level navigation architecture that contains a topological-metric memory structure and a deep image-based controller.
    Authors: Shaojun Cai; Yingjia Wan   
    Keywords:  Autonomous driving, visual navigation, topological map, imitation learning, Carla simulator
Frid Jun19  
All day
Favorite
Teaser picture for paper
    Authors: James Hays   
Frid Jun19  
9:00 AM - 9:30 AM
Favorite
Teaser picture for paper
    Authors: Fisher Yu   
Frid Jun19  
9:30 AM - 10:00 AM
Favorite
Teaser picture for paper
    Authors: Deva Ramanan, Drago Anguelov, Emilio Frazzoli, ...   
Frid Jun19  
10:00 AM - 11:30 AM
Favorite
Teaser picture for paper
    Authors: James Hays   
Frid Jun19  
9:00 PM - 9:30 PM
Favorite
Teaser picture for paper
    Authors: Fisher Yu   
Frid Jun19  
9:30 PM - 10:00 PM
Favorite
Teaser picture for paper
Frid Jun19  
10:00 PM - 11:30 PM
Favorite