Deep Learning for Semantic Visual Navigation
June 16 2019
Long Beach Convention Center
Hall 103B
in conjunction with CVPR 2019
Visual navigation, the ability of an autonomous agent to find its way in a large, visually complex environment based on visual information, is a fundamental problem in computer vision and robotics.
This workshop provides a forum to promising ideas proposing to advance visual navigation by combining recent developments in deep and reinforcement learning. A special focus lies on approaches that incorporate more semantic information into navigation, and combine visual input with other modalities such as language.
Topics
Deep Reinforcement Learning for Navigation
Deep Learning for Planning and Reasoning
Representations for Navigation
Vision-Based, Semantic SLAM
Learning methods for SLAM
3D Image Understanding
Simulation Environments for Navigation and Control
Language and Robot Navigation
Object Detection, Segmentation for Robot Navigation
June 16, 2019
Long Beach Convention Center
Hall 103B
Schedule
08:40- 08:45 Welcoming
8:45 - 09:20 Invited Talk:
Georgia Gkioxari, FAIR,
09:20 - 10:00 Invited Talk:
Jana Kosecka, GMU
Generalizable Visual Representations for Navigation and Visuomotor Control
10:00 - 10:45 Poster session + Refreshments
10:45 - 11:20
Ruslan Salakhutdinov, CMU
Action Navigation Networks -- Winner of Habitat Challenge
11:20 - 12:00 Invited Talk:
Amir Zamir, Stanford University,
Learning to Navigate using Mid-level Visual Priors: Enhanced Generalization and Sample Efficiency
12:00 - 13:30 Lunch
13:30 - 14:05 Brandon Rothrock
14:05 - 14:45
Vladlen Koltun, Intel
Does Computer Vision Matter for Science?
14:45 - 15:15 Poster Session
15:15 - 15:55 Invited Talk:
Jitendra Malik, UC Berkeley and FAIR,
Visually Guided Navigation - the Role of Mapping, Landmarks and Learning
15:55 - 16:30 Panel Discussion
Participants: Jana Kosecka, Russ Salakhudinov, Andy Davison, Brandon Rothrock, Amir Zamir
Papers
Learning to Follow Directions in Street View, Karl Moritz Hermann, Mateusz Malinowski, Piotr Mirowski, Andras Banki-Horvath, Keith Anderson, Raia Hadsell
Robustness Meets Deep Learning: An End-to-End Hybrid Pipeline for Unsupervised Learning of Egomotion, Alex Zihao Zhu, Wenxin Liu, Ziyun Wang, Vijay Kumar, Kostas Daniilidis
Gibson Env V2: Embodied Simulation Environments for Interactive Navigation, Fei Xia, Chengshu Li, Kevin Chen, Roberto Martin-Martin, Noriaki Hirose, Amir R. Zamir, Silvio Savarese
Combining Optimal Control and Learning for Visual Navigation in Novel Environments, Somil Bansal, Varun Tolani, Saurabh Gupta, Jitendra Malik, Claire Tomlin
Visual Landmark Selection for Generating Grounded and Interpretable Navigation Instructions, Sanyam Agarwal, Devi Parikh, Dhruv Batra, Peter Anderson, Stefan Lee
Two Body Problem: Collaborative Visual Task Completion, Unnat Jain, Luca Weihs, Eric Kolve, Mohammad Rastegari, Svetlana Lazebnik, Ali Farhadi, Alexander Schwing, Aniruddha Kembhavi
Scene Motion Decomposition for Learnable Visual Odometry, Igor Slinko, Anna Vorontsova, Filipp Konokhov, Olga Barinova, Anton Konushin
Are You Looking? Grounding to Multiple Modalities in Vision-and-Language Navigation, Ronghang Hu, Daniel Fried, Anna Rohrbach, Dan Klein, Trevor Darrell, Kate Saenko
Viewpoint Optimization for Autonomous Strawberry Harvesting with Deep Reinforcement learning, Jonathon Sather, Xiaozheng Jane Zhang
Learning Navigation Subroutines by Watching Videos, Ashish Kumar, Saurabh Gupta, Jitendra Malik
Speakers
Jitendra Malik
UC Berkeley & FAIR
Vladlen Koltun
Intel Labs
Jana Kosecka
George Mason University
Ruslan Salakhutdinov
CMU
Georgia Gkioxari
FAIR
Brandon Rothrock
JPL
Amir Zamir
Stanford University
Submission Details
The workshop accepts 2 page abstracts up to 8 page full papers. The paper should be formatted the same way a conference submission to CVPR is formatted.
Each paper will be reviewed by at least two reviewers. The workshop accepts double submissions. The reviewing process is single blind, i.e. the reviewer will know the authors but the other way around.
The accepted papers will be linked online on the workshop website. They will not be part of the CVPR proceedings or otherwise archived.
The submission webpage is under: https://cmt3.research.microsoft.com/SEMNAV2019, and the deadline is March 25th.
Organizers
Niko Sünderhauf
Australian Center for Robotic Vision & Queensland University of Technology
Program Committee
Alexey Dosovitskiy, Intel Labs,
Angel Xuan Chang, Eloquent Labs
Arsalan Mousavian, NVIDIA
Daniel Gordon, University of Washington
Gim Hee Lee, National University Singapore
Honglak Lee, University of Michigan and Google Research
Manolis Savva, Simon Fraser University
Piotr Mirowski, DeepMind
Saurabh Gupta, FAIR / UIUC
Zeeshan Zia, Microsoft