Dario Cadena Lerma
Autonomous Systems Lab
Institute of Robotics and Intelligent Systems
Dep. of Mechanical and Process Engineering
Office: J-201, LEE
Leonhardstrasse 21, Zurich, Switzerland 8092
Phone: (+41) (0)44 633-8803
dot lerma at gmail dot com
Currently, I'm senior researcher at the Institute of Robotics and
Systems at ETH Zurich, working with the Autonomous Systems Lab. lead by
Roland Siegwart. I'm also a research affiliate in the ARC Centre of
Excellence for Robotic Vision.
Previously, I was a Senior Researcher at School of Computer Science of
The University of Adelaide, working with Ian Reid,
and a Postdoc at CS Department of George Mason University, working with
Jana Kosecka. I got my PhD, under the supervision of Jose Neira, in the
Department of Computer Science and System Engineering at the University
My main research interest lies in the interception of perception and
learning in robotics. I am particularly interested on how to provide
machines the capability of understanding this ever changing world
through the sensory information they can gather. I have work
intensively on Robotic Scene Understanding, both geometry and
covering Semantic Mapping, Data Association and Place Recognition
tasks, Simultaneous Localization and Mapping problems, as well as
persistent mapping in dynamic environments.
- Jul. 11/19: Our journal
paper "SegMap: Segment-based Mapping and Localization using Data-driven
Descriptors" is now published in the International
Journal of Robotics Research (IJRR).
- Jun. 15/19: Three papers
accepted to IROS'19:
- Free-Space Features:
Global Localization in 2D Laser SLAM Using Distance Function Maps, by
- Flexible Trinocular:
Constrained, Non-Rigid, Wide-Baseline, Relative Visual-Inertial
Multi-View Stereo Dense Reconstruction for High Speed UAV Navigation
and Map Building, by Timo Hinzmann.
- OREOS: Oriented
Recognition of 3D Point Clouds in Outdoor Scenarios, by Lukas Schaupp.
- Jun. 15/19: Our paper "Airborne Particle Classification in LiDAR PointClouds Using Deep Learning" has been accepted to FSR'19. This is a result of our frutiful collaboration with QUT.
- Jun. 10/19: Two papers
accepted to RA-L:
- Multiple Hypothesis
Semantic Mapping for Robust Data Association, by Lukas Bernreiter.
Instance-Aware Semantic Mapping and 3D Object Discovery, by Margarita
- Jun. 06/19: The survey
paper "The Current State and Future Outlook of Rescue Robotics" has
been accepted for publication in the Journal of Field
Robotics. This work was led by Jeff Delmerico.
- Apr. 12/19: Our paper
"VIZARD: Reliable Visual Localization for Autonomous Vehicles in Urban
Outdoor Environments" has been accepted to IV 2019.
- Apr. 01/19: Our journal
paper "Observability-Aware Self-Calibration of Visual and Inertial
Sensors for Ego-Motion Estimation" is now published in the IEEE
- Mar. 12/19: Our paper
"From Coarse to Fine: Robust Hierarchical Localization at Large Scale"
been accepted to CVPR'19.
- Mar. 09/19: Our journal
"Appearance‐based Landmark Selection for Visual Localization" has been
accepted for publication in the Journal
of Field Robotics.
- Feb. 15/19: Three papers
accepted to ICRA'19:
- An Approach for Semantic
Segmentation of Tree-Like Vegetation, by Teja, Lukas and Giuseppe.
- Object Classification
Based on Unsupervised Learned Multi-Modal Features for Overcoming
Sensor Failures, by Julia Nitsch.
- Empty Cities: Image
Inpainting for a Dynamic-Object-Invariant Space, by Berta Bescos.
- Feb. 01/19: Two papers
accepted to RA-L:
- Where Should I Walk?
Predicting Terrain Properties from Images via Self-supervised Learning,
by Lorenz Wellhausen.
- Experimental Comparison
of Visual-Aided Odometry Methods for Rail Vehicles, by Florian Tschopp.
- Jan. 01/19: The journal
paper "3D multi-robot patrolling with a two-level coordination
strategy" is now published in Autonomous
Robots. This was a fruitful collaboration led by Luigi
- Nov. 01/18: Julia Nitsch
will be presenting our paper
"3D Ground Point Classification for Automotive Scenarios" at
- Sep. 01/18: I have been
Editor for RA-L
in the area of Visual Perception and Learning.
- Sep. 01/18: Our paper
"Leveraging Deep Visual Descriptors for Hierarchical Efficient
been accepted to CoRL'18.
- Aug. 24/18: Two papers
accepted to RA-L:
- Automatic Segmentation
Structure from Point Cloud Data, by Teja Digumarti.
Imitation: Sample Efficient Deep Reinforcement Learning for Map-less
Navigation by Leveraging Prior Demonstrations, by Mark Pfeiffer.
- Jul. 01/18: Two papers
accepted to IROS'18:
- Modular Sensor Fusion
Segmentation, by Hermann Blum.
- C-Blox: A Scalable and
TSDF-Based Dense Mapping Approach, by Alex Millane.
- May. 15/18: Our paper
Segment Mapping using Data-Driven Descriptors" has
been accepted to RSS'18.
- Feb. 01/18: Three papers
accepted to RA-L:
- X-View: Graph Based
Semantic Multi-View Localization, by Abel Gawel.
Segment-Based Localization in 3D Point Clouds, by Renaud Dubé.
- Free LSD: Prior-Free
Visual Landing Site Detection for Autonomous Planes, by Timo Hinzmann.
- Jan. 15/18: Our paper "A
Model for Interaction-Aware Pedestrian Motion Prediction in Object
Cluttered Environments" has
been accepted to ICRA'18.
- Aug. 15/17: Two papers
accepted to SSRR'17:
- 3D Registration of
Aerial and Ground
Robots for Disaster Response: An Evaluation of Features, Descriptors,
and Transformation Estimation, by Abel Gawel.
- Reliable Real-time
Change Detection and Mapping for 3D LiDARs, by Lorenz Wellhausen.
- June 15/17: Our paper "An
Online Multi-Robot SLAM System for 3D LiDARs" has
been accepted to IROS'17.
- Feb. 01/17: Three papers
accepted to ICRA'17:
- TSDF-based Change
Detection for Consistent Long-Term Dense Reconstruction and Dynamic
Object Discovery, by Marius Fehr.
Perception to Decision: A Data-driven Approach to End-to-end Motion
Planning for Autonomous Ground Robots, by Mark Pfeiffer.
- SegMatch: Segment based
place recognition for 3D point clouds, by Renaud Dube.
- Dec. 01/16: Our position
paper on the future challenges in SLAM has been accepted for
publication as a survey paper in the Transaction on RObotics (available
- Sep. 01/16: I have been
appointed as a Research Affiliate in the ARC Centre of Excellence
Robotic Vision. It is a great honor to have the opportunity
collaborating with the amazing talented people at the ACRV.
- July 10/16: Our paper on
Measuring the Performance of Single Image Depth Estimation Methods has
been accepted to IROS'16.
- July 1/16: I will continue
serving for one more year as Associate Editor for the IEEE
Robotics and Automation Letters.
- May 18/16: Our paper on
Multi-modal Auto-Encoders as Joint Estimators for Robotics Scene
Understanding has been accepted to RSS 2016.
- March 26/16: I'm
co-organizing the workshop Geometry and
Beyond - Representations, Physics, and Scene Understanding for Robotics
in conjunction with
RSS 2016 .
- March 01/16: Our paper on
Direct Semi-dense SLAM for Rolling Shutter Cameras has been accepted
- Jan. 11/16: I have joined
the Autonomous Systems Lab. at ETH Zurich.
- Dec. 31/15: I have
finished an amazing time working in Australia with Ian Reid, the ACVT
and the Australian Center for Robotic Vision.
- May 01/15: I'm
co-organizing the workshop The Problem
of Mobile Sensors: Setting future goals and indicators of progress for
in conjunction with
RSS 2015 .
- March 01/15: Our paper on
A Fast, Modular Scene Understanding System using Context-Aware
Object Detection has been accepted to ICRA'15.
- Aug. 07/14: Our paper on
Semantic Parsing for Priming Object Detection in Indoors RGB-D Scenes
has been accepted to the IJRR's special issue on Robot Vision.
- Aug. 07/14: Yasir's paper
on comparisons on robust back-ends has
been accepted to IROS'14
- Mar. 25/14: I will be
chairing the session on Computer
Vision: Semantic Scene Understanding
- Jan. 14/14: Our
paper on semantic segmentation with heterogeneous sensor coverages has
been accepted to the ICRA'14
- Nov. 01/13: I have joined
the Australian Center for Visual Technologies in Adelaide, Australia,
to work with Ian Reid.
- Oct. 07/13: I
gave a GRASP
Special Seminar at
University of Pennsylvania, Philadelphia, PA.
- Oct. 03/13: I
gave a talk at the CS
Washington University, Washington, DC.
- Sep. 25/13: Our
paper on recursive inference for semantic
segmentation has been accepted to the ISRR'13
- Sep. 20/13: I have
finished a great time working with Jana
Kosecka, now I'm in the process of moving to Adelaide, to work with Ian