Robuddies Reading Group
This is the web page of the LCI reading group Robuddies.
Robuddies meets Fridays at 1:00pm (for 2014 Winter Term 2) in ICCS 146
The topics for Robuddies are Robotics and Computer Vision.
Each week someone presents a paper/tutorial or their current
research, or some mix of these.
A schedule of upcoming papers can be found below under the heading Upcoming
Papers. Under the heading Finished Papers is a log of earlier
topics. If you want to be added to the schedule, please contact the Robuddies
Czar and Czarina, who currently happen to be Neil Traft (ntraft AT cs ubc ca)
and Lili Meng (lilimeng1103 AT gmail com).
We have two mailing lists.
To subscribe to the appropriate mailing list, send a message to "majordomo[at]cs.ubc.ca" with the words
"subscribe robuddies-seminars" or "subscribe robuddies" in the body. Note that robuddies-seminars is a
moderated mailing list so non-members cannot mail to it. Contact the Czars if you have trouble registering.
- firstname.lastname@example.org is the place to get information on upcoming seminars.
- email@example.com is a list for interesting news and tidbits about robotics and computer vision. You're welcome to join our conversations whether or not you can attend our meetings!
The picture above is a panorama of our lab. Click on it for a Quicktime VR tour. (shot by Kenji)
Upcoming Presentation Schedule
- April 24, 2015: Neil Traft will introduce the RRT* algorithm: Sertac Karaman and Emilio Frazzoli. Sampling-based algorithms for optimal motion planning. The International Journal of Robotics Research, 2011. [PDF]
- May 01, 2015: This could be you!
Finished Papers, 2015
- March 27, 2015: Neil Traft presented some of his work with Ian Mitchell on motion planning under uncertainty, toward synthesizing a smooth, approximately optimal control given imperfect state information.
- March 20, 2015: In a joint meeting with CVRG, Fred Tung presented Abhijit Kundu, Yin Li, Frank Daellert, Fuxin Li and James M. Rehg. Joint Semantic Segmentation and 3D Reconstruction from Monocular Video. Computer Vision—ECCV 2014. [PDF]
- March 13, 2015: This week we have a visitor, Anqi Xu from Greg Dudek's Mobile Robotics Lab at McGill! He's going to tell us about some of their recent research into improving the efficiency of human-robot systems by building trust in the relationship: Towards Efficient Collaborations with Trust-Seeking Adaptive Robots.
- March 6, 2015: Victor Gan presented a recent variant of RRTs that biases exploration away from obstacles: Jory Denny, Evan Greco, Shawna L. Thomas, Nancy M. Amato. MARRT: Medial Axis Biased Rapidly-Exploring Random Trees. IEEE Intl. Conf. on Robotics and Automation (ICRA). Hong Kong, China, 2014. [PDF]
- February 23, 2015: Lili Meng presented an ICRA 2015 paper the integration of camera and lidar for odometry: J. Zhang and S. Singh. Visual-lidar Odometry and Mapping: Low-drift, Robust, and Fast. IEEE Intl. Conf. on Robotics and Automation (ICRA). Seattle, WA, May 2015. [PDF] [Video 1] [Video 2]
- February 6, 2015: Ankur Gupta and Neil Traft presented a technical paper on the RoboBrain architecture from a high-level view: Ashutosh Saxena, et al. RoboBrain: Large-Scale Knowledge Engine for Robots. arXiv preprint, 2014. [PDF]
- January 23, 2015: Julieta Martinez presented one of the more interesting papers from WACV, about using monocular cameras for self-driving vehicles: Yao, Jian, Srikumar Ramalingam, Yuichi Taguchi, Yohei Miki, and Raquel Urtasun. Estimating Drivable Collision-Free Space from Monocular Video. [PDF]
- January 16, 2015: Justin Hart outlined for us the work he did for his PhD thesis. Justin recently defended his dissertation on Robot Self-Modeling at Yale under the supervision of Brian Scassellati. Descriptions of his work have appeared in New Scientist, BBC News, and numerous other publications.
- January 5, 2015: Bikram Adhikari gave his Masters thesis presentation. The title of Bikram's talk was A Single Subject Participatory Action Design Method for Powered Wheelchairs Providing Automated Back-in Parking Assistance to Cognitively Impaired Older Adults: A pilot study.
Finished Papers, 2014
- December 9, 2014: Lili Meng told us about her ongoing trials and tribulations with SLAM and odometry, and her usage of the following algorithm: A Sliding-Window Visual-IMU Odometer Based on Tri-focal Tensor Geometry by Jwu-Sheng Hu and Ming-Yuan Chen In Proc. of the IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, China, 2014. [PDF] [Matlab code]
- November 25, 2014: Neil Traft talked about a proposed method for benchmarking navigation algorithms. (But it turned out to be rather contrived and doesn't seem like it will gain much traction.) Christoph Sprunk, Jorg Rowekamper, Gershon Parent, Luciano Spinello, Gian Diego Tipaldi, Wolfram Burgard, and Mihai Jalobeanu. An Experimental Protocol for Benchmarking Robotic Indoor Navigation. In Proc. of the International Symposium on Experimental Robotics (ISER). 2014. [PDF][website]
- November 18, 2014: Lili Meng told us about the 3D SLAM problem she's been working on lately. She spoke about appearance-based mapping and the open source toolbox openFABMAP. Here's the paper: Mark Cummins and Paul Newman. FAB-MAP: Probabilistic Localization and Mapping in the Space of Appearance. The International Journal of Robotics Research 27.6 (2008): 647-665. [PDF]
- November 4, 2014: Neil Traft told us about some work out of Wolfram Burgard's lab on predicting the movement of pedestrians: H. Kretzschmar, M. Kuderer, and W. Burgard. Learning to Predict Trajectories of Cooperatively Navigating Agents. In Proc. of the IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, China, 2014. [PDF]
- October 28, 2014: Victor Gan presented joint work between CMU and JPL from RSS 2014, which cleverly incorporated both global and local objectives into a single optimization problem: Desaraju, V. R., Michael, N., Humenberger, M., Brockers, R., Weiss, S., & Matthies, L. Vision-based Landing Site Evaluation and Trajectory Generation Toward Rooftop Landing. In Proceedings of Robotics: Science and Systems. July, 2014. Berkeley, USA. [PDF]
- October 22, 2014: Determining Proper Grasp Configurations for Handovers through Observation of Object Movement Patterns and Inter-object Interactions during Usage — During a special session held jointly with CARIS Lab, we heard from recent UBC alumnus Wesley Chan, now a PhD candidate in the JSK Lab at the University of Tokyo, about his work on human-robot handovers.
- October 14, 2014: Jim Little brought us up to date on the latest techniques for performing robust visual odometry on cheap mass-market cell phone cameras: the same technology employed in Google's Project Tango phone. The paper was Guo, Chao X., Dimitrios G. Kottas, Ryan C. DuToit, Ahmed Ahmed, Ruipeng Li, and Stergios I. Roumeliotis. Efficient Visual-Inertial Navigation using a Rolling-Shutter Camera with Inaccurate Timestamps. Proceedings of Robotics: Science and Systems, Berkeley, USA, July, 2014. [PDF]
- September 30, 2014: Lili Meng presented this paper from IROS 2014: Real-time Depth Enhanced Monocular Odometry Ji Zhang, Michael Kaess, and Sanjiv Singh. [PDF]
- September 16, 2014: Neil Traft presented an interesting robotic navigation system which takes inspiration from what neuroscience tells us about rodent spatial memories: Milford, M. J., Wiles, J., & Wyeth, G. F. (2010). Solving Navigational Uncertainty Using Grid Cells on Robots. PLoS Computational Biology, 6(11). [paper, slides]
- August 24, 2014: Tricia Jose told us about her work as a USRA for the Canwheel project. Tricia has been examining the data from the Canwheel Wizard of Oz study and working on a new remote control system for the wheelchair.
- August 16, 2014: Neil Traft presented this moving object tracking paper from the recent RSS: Combining 3D Shape, Color, and Motion for Robust Anytime Tracking. David Held, Jesse Levinson, Sebastian Thrun, and Silvio Savarese. [PDF]
- August 11, 2014: Ian Mitchell and Lili Meng will give a joint talk on RSS 2014.
- July 28, 2014: Umang Gupta presented his summer work at UBC as a Mitacs Globalink intern. Umang is a Masters student in Information and Communication Technology at IIT Delhi, and has been working on various UIs to ease the task of collecting ground truth annotations for human pose data. Neil Traft also updated us on the progress of the CARIS PR2 summer hackathon.
- July 14, 2014: Ankur Gupta (and by proxy, Julieta Martinez) gave a conference report on their visit to CVPR 2014 in Columbus, Ohio.
- July 7, 2014: Alireza Shafaei led the discussion on the paper titled "Learning Everything about Anything: Webly-Supervised Visual Concept Learning", published in CVPR 2014. The paper can be found here.
- June 30, 2014: Junaed Sattar led the discussion on the ICRA 2014 paper titled "Shady Dealings: Robust, Long- Term Visual Localisation using Illumination Invariance", by Colin McManus, Winston Churchill, Will Maddern, Alex Stewart and Paul Newman from the Oxford Mobile Robotics Group. The paper can be found here.
- June 9, 2014: Junaed Sattar gave a trip report on the recently concluded ICRA 2014 conference, held in Hong Kong.
- June 2, 2014: Damon Miller, Carolyn Shen, and Tricia Jose talked about their summer projects at the collaborative robotics lab (otherwise known as room x209!). All three are summer USRAs, working under the supervision of Professor Ian Mitchell, and are working on a number of interesting projects.
- May 26, 2014: Neil Traft demonstrated the Google Project Tango phone to the group, its capabilities, and what cool things we can do with it.
- May 7, 2014: Ankur Gupta presented the paper titled "Reconstructing PASCAL VOC" by Sara Vicente, Joao Carreira, Lourdes Agapito, Jorge Batista. This is a CVPR 2014 preprint and can be found here.
- April 23, 2014: Anil Mahmud, MSc student being supervised by Professor Woodham, gave his master's talk, titled "Photometric Stereo: Applications and Recent Advances"
- April 16, 2014: Lili Meng, PhD candidate at UBC ECE presented the paper titled "Robust Real-Time Visual Odometry for Dense RGB-D Mapping" by Thomas Whelan, Hordur Johannsson, Michael Kaess, John J. Leonard and John McDonald, a paper that came out in ICRA 2013. It can be found here.
- April 9, 2014: Julieta Martinez and Gregor Miller gave a report on their recent trip to the WACV 2014 conference.
- April 2, 2014: Lili Meng, PhD candidate at UBC ECE presented the paper titled "Change Their Perception: RGB-D for 3-D Modeling and Recognition" by Xiaofeng Ren, Dieter Fox, and Kurt Konolige, from the IEEE Robotics and Automation Magazine (RAM), 2013. The paper can be found on page 49 of this link.
- March 19, 2014: Neil Traft presented a paper on robot navigation in a crowd, based on papers by Trautman et al., which can be found here (IROS2010) and here (ICRA2013).
- February 26, 2014:
Sancho McCann gave a practice talk of his then-upcomong PhD defense.
- February 12, 2014:
Anil Mahmud presented the following paper for our next Robuddies seminar: "Photometric Stereo With Non-Parametric and Spatially-Varying Reflectance". ALdrein, Zickler and Kriegman.
- February 5, 2014:
Victor Gan led us through the discussion of the ICRA 2013 best conference paper award recipient, "Control Design along Trajectories with Sums of Squares Programming", by Anirudha Majumdar, Amir Ali Ahmadi, Russ Tedrake.
- January 29, 2014:
Bikram Adhikari, MSc candidate with Professors Alan Mackworth and Ian Mitchell, talked about his recent experience with the Wizard-of-Oz (WoO) trials, conducted as part of a series of research trials under the CanWheel (www.canwheel.ca) project. These trials were conducted to identify possible policies for collaborative control of smart wheelchairs. Bikram discussed the background, details of the trials, challenges and future directions in his talk.
- January 22, 2014:
Anil Mahmud presented the paper titled "Calibrating Photometric Stereo by Holistic Reflectance Symmetry Analysis".
- January 15, 2014:
We had a joint presentation by Fred Tung and Julieta Martinez. They discussed last year's ICCV Marr prize winning paper, "From Large Scale Image Categorization to Entry-Level Categories." The authors are Vicente Ordonez, Alexander C. Berg, and Tamara L. Berg from the University of North Carolina at Chapel Hill, Jia Deng from Stanford University, and Yejin Choi from Stony Brook University. Paper can be found here.
Finished Papers, 2013
- December 9, 2013:
There was an ICCV2013 roundtable discussion for this Robuddies seminar. Ankur suggested sharing a list of papers on a common document, and limit speakers to a maximum of 10 minutes (depending on how many we have). All the papers from ICCV are available here.. List of presenters and their papers are listed in this google drive document (might not be publicly accessible).
- December 2, 2013: Julieta Martinez presented the paper titled "To aggregate or not to aggregate: Selective match kernels for image search", by Giorgos Tolias, Yannis Avrithis (NTUA), Herv Jgou (INRIA). This is an ICCV 2013 paper and can be found here.
- November 4, 2013: Devon Ash from the UBC ThunderBots RoboCup team presented the team, their hardware and the systems aspect in general.
- October 28, 2013: Daniel Troniak from CS@UBC gave a presentation on his masters research in dexterous manipulation.
- October 21, 2013: Professor Richard Vaughn and his group from Simon Fraser University gave two short talks in preperation for their talks at the upcoming IROS 2013 conference. Talks were given by Mani Moajjemi and Shokoofeh PourMehr, both PhD students at SFU.
- October 7, 2013: Ankur Gupta presented the paper titled "Unsupervised Joint Object Discovery and Segmentation in Internet Images" by Rubinstein at el. This is a CVPR 2013 paper, and can be found here.
- September 30, 2013: This was a special showcase session of Robuddies highlighting current research in the lab in vision. robotics and AI. Professors Little and Mitchell briefly described some of the projects in vision and assistive robotics, respectively. Students Ankur Gupta and Sancho McCann talked their own research in motion tracking and object recognition. Junaed Sattar rounded up the hour by briefly describing multiple projects he has been involved with in HRI, robotics, and ubiquitous HRI.
- September 23, 2013: Junaed Sattar presented the paper titled "Anticipating Human Activities using Object Affordances for Reactive Robotic Response", by Hema Koppula and Ashutosh Saxena. This was published in Robotics: Science and Systems IX, June 2013 and was the recipient of the best student paper. The paper can be found here.
- September 16, 2013: Ajung Moon, PhD student with Professort Elizabeth Croft in the CARIS lab and her fellow "hackathon-mates" gave us a talk on their latest human study on the effect of gaze during robot-to-human handovers. This study took place in ICCS146, coincidentally, on Imagine Day (September 2).
- July 31, 2013: Professor Frank Park from Seoul National University gave a talk titled "Differential Geometric Algorithms for Vision and Image Processing". This was the last Robuddies for the summer session.
- July 23, 2013: Junaed Sattar presented the paper titled "Sorry Dave, I’m Afraid I Can’t Do That: Explaining Unachievable Robot Tasks Using Natural Language", by Raman et al, which was published in RSS (Robotics: Science and Systems IX) 2013. The paper can be found here.
- July 16, 2013: Ankur Gupta presented the paper titled "Intrinsic Scene Properties from a Single RGB-D Image", by Jonathan T. Barron and Jitendra Malik, at UC Berkeley. The paper can be found here.
- July 8, 2013: David Meger gave a practice talk for his PhD defense. As a followup, he successfully defended on the 9th of July.
- July 2, 2013: A CVPR 2013 conference summary was presented by Ankur Gupta, Sancho McCann, Fred Tung and Julieta Martinez.
- June 18, 2013:
Sancho McCann, PhD candidate with Dr. David Lowe, gave a talk on a CVPR2013 paper titled "Fast, Accurate Detection of 100,000 Object Classes on a Single Machine", by Thomas Dean, Mark A. Ruzon, Mark Segal, Jonathon Shlens, Sudheendra Vijayanarasimhan and Jay Yagnik, from Google. The PDF is here.
- June 11, 2013:
Fred Tung, PhD Candidate with Dr. Jim Little, gave a talk on a TPAMI 2012 paper titled "Measuring the objectness of image windows", by Bogdan Alexe, Thomas Deselaers, and Vittorio Ferrari. PDF can be found here.
- June 4, 2013:
Luz Abril Torres Mendez, visiting Professor from CINVESTAV, Saltillo, Mexico gave a talk on the research of her group, along with visiting masters student Luis Gonzales.
- May 28, 2013:
Ian Dewancker, Masters student with Dr. Ian Mitchell, gave a talk on his research on lifespace mapping and MobiSense.
- May 15, 2013:
Junaed Sattar gave a trip and conference report on the recently concluded ICRA conference in Karlsruhe, Germany. The slides can be found on Google Docs.
- March 19, 2013:
Professor Gregory Dudek gave a talk titled "Robot teams to assist humans in scientific discovery". Dr. Dudek is a Professor and Chair at the McGill School of Computer Science, and also the founding Director of the NSERC Canadian Field Robotics Network (NCFRN). The video of his talk can be found here.
- March 11, 2013:
Mauricio Brauer, visiting Master's student from Mexico City, from the Department of Computer Science, Artificial Intelligence Laboratory, National Polytechnic Institute (IPN) gave a talk on his research. He is pursuing a master's degree in robotics, with a focus on humanoid robots. He is visiting the LCI as a guest of Jim Little, and stayed till the end of April.
- March 4, 2013:
Professor Yuji Iwahori, visiting us from Chubu University in Japan, as a guest of Professor Bob Woodham, presented his research on 3D Shape Recovery and Object Detection in Computer Vision.
- February 4, 2013: Bikram Adhikari, Masters candidate with Professor Alan Mackworth, presented the paper on Wheelchair navigation, titled "Wheelchair Navigation Assistance in the FP7 Project RADHAR: Objectives and Current State", by Eric Demeester, Emmanuel Vander Poorten, Alexander Huntemann and Joris De Schutter. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), workshop on Navigation and Manipulation Assistance for Robotic Wheelchairs, October 2012, Vilamoura, Portugal. The paper can be found here.
- January 21, 2013: Julieta Martinez presented the paper by Jianxiong Xiao and Yasutaka Furukawa, "Reconstructing the World's Museums". European Conference on Computer Vision (EECV 2012). Firenze, Italy. October 2012. The authors have set up a nice website with a running demo here.
- January 14, 2013: Junaed Sattar presented a paper on robots adapting to appearance changes during long-term autonomy, by Winston Churchill and Paul Newman. "Practice Makes Perfect? Managing and Leveraging Visual Experiences for Lifelong Navigation". IEEE International Conference on Robotics and Automation (ICRA). Minnesota, USA. May 2012.
- January 7, 2013: Ankur Gupta presented a tch report from MIT, on "What do HOGs really see?" and discused inverse HOGs.
Finished Papers, 2012
- December 10, 2012: Georgii Oleinikov presented the CVPR 2012 best student paper, "A Simple Prior-free Method for Non-Rigid Structure-from-Motion Factorization". Yuchao Dai, Hongdong Li, Mingyi He.
- December 3, 2012: Sancho McCann gave us a conference report on ACCV 2012, and talked about these two papers from the conference: VISOR: Towards On-the-Fly Large-Scale Object Category Retrieval. K. Chatfield, A. Zisserman. and Detecting Partially Occluded Objects with an Implicit Shape Model Random Field. Paul Wohlhart, Michael Donoser, Peter M. Roth, and Horst Bischof. Winner of the Best Paper Award.
- November 26, 2012: David Lowe presented "Unsupervised Discovery of Mid-Level Discriminative Patches," Saurabh Singh, Abhinav Gupta, Alexei A. Efros. A nice application is given in the following paper at SIGGRAPH 2012, which David briefly discussed: "What makes Paris look like Paris?" Carl Doersch, Saurabh Singh, Abhinav Gupta, Josef Sivic, Alexei A. Efros.
- November 5, 2012: Ian Dewancker, MSc student in CS supervised by Ian Mitchell, gave a talk on using Android phones for place recognition, by using machine learning on visual and WiFi signal data.
- October 29, 2012: Junaed Sattar presented a paper on information theoretic human-robot dialogs, by Stefanie Tellex et al. "Toward Information Theoretic Human-Robot Dialog", Stefanie Tellex, Pratiksha Thaker, Robin Deits, Thomas Kollar, Nicholas Roy.
- October 22, 2012: Guest speaker Dr. Kevin Cannons, MITACS Post-Doctoral Fellow from SFU, gave a talk on his PhD work at York, with Rick Wildes. "Visual Tracking and Action Spotting using Spatiotemporal Oriented Energies".
- October 15, 2012: Chen Xing gave a talk on his Masters work, titled "Capturing Fast Motion with Consumer Grade Unsynchronized Rolling-Shutter Cameras". Chen was co-supervised by Bob Woodham and Wolfgang Heidrich.
- October 1, 2012: Koosha Khalvati gave a practice talk for his IROS'12 paper titled "Active Robot Localization with Macro Actions", with Alan Mackworth as the co-author.
- September 24th, 2012: Ankur Gupta presented ActionBanks, for activity analysis in videos. Action Bank™: A High-Level Representation of Activity in Video, CVPR'12
- June 29th, 2012: Sancho McCann and Marius Muja present a selection of highlights from their recent trip to the CVPR conference in Providence, RI.
- June 15th, 2012: Daniel Troniak presents work done by a group of students at UBC on making Charlie, the PR2 robot, ride the elevator between floors in the ICICS building.
- June 8th, 2012: Ian Dewancker presents with work on reinforcement learning for an autonomous Connect-4 playing robot.
- June 1st, 2012: Visiting talk by Saul Greenberg from U of Calgary on HCI.
- May 11th, 2012: Dave Meger practices the presentation for Parnian Alimi's thesis work that will appear at ICRA 2012 SPME workshop.
- May 3rd, 2012 3:30-5 (note special Thursday time): Visiting talk by Sven Dickinson.
- April 27th, 2012: Dave Meger will present: R. Girshick, P. Felzenszwalb, D. McAllester. "Object Detection with Grammar Models". NIPS 2011. [External Link]
- April 20th, 2012: Visiting talk by Chris Williams, University of Edinburgh. Modelling of visual textures and object parts.
- April 13th, 2012: Parnian Alimi MSc presentation: Object persistence in 3D for home robotics.
- March 30th, 2012: Machiel Van der Loos talks about the history and trends of therapy robotics.
- March 9th, 2012: Maodi Hu presents "Learning Spatiotemporal Graphs of Human Activities", William Brendel and Sinisa Todorovic, ICCV 2011. [External Link].
- March 2nd, 2012: Junaed Sattar presents Learning the semantics of object-action relations by observation. Eren Erdal Aksoy, Alexey Abramov, Johannes Dörr, Kejun Ning, Babette Dellen, and Florentin Wörgötter. [External Link].
- Feb 24th, 2012: Visitor Richard Zemel talks about his work on recommender systems, active learning and solving the problem of matching papers to reviewers.
- Feb 10th, 2012: Ankur Gupta presents Secrets of Optical Flow Estimation and Their Principles. Sun, D, Roth, S, and Black M, J. CVPR 2010. [External Link].
- Feb 3rd, 2012: Visiting professor James Clark presents his work on Attention Tracking. The supporting papers will be [External Link] and [External Link].
- Friday, Jan 27th, 2012: No regular robuddies due to PhD Thesis defense by Kenjii Okuma.
- Friday, Jan 20th, 2012: Martin Lau will present his BSc thesis work on Robust Methods for Automatic Determination of Puck Possession and Location.
Friday, Jan 13th, 2012: Junaed Sattar talks about his PhD thesis on visual human robot interaction.
Thursday, Jan 5th, 2012: Shervin Tari presents his MSc work on Automatic initialization for broadcast sports videos rectification.
Finished Papers 2011
A log of 2011 papers can be found here.
Finished Papers 2010
A log of 2010 papers can be found here.
Finished Papers 2009
A log of 2009 papers can be found here.
Finished Papers 2008
A log of 2008 papers can be found here.
Finished Papers 2007
A log of 2007 papers can be found here.
Finished Papers 2006
A log of 2006 papers can be found here.
Curious George Page
The UBC LCI robot that won the Semantic Robot Vision Challenge is called
Curious George. Find out more about this robot here.
Historic Robuddies Page
The robot partners page can be found here. This is
where the name Robuddies comes from.