Place categorization and semantic mapping on a mobile robot
By Niko Sünderhauf, Feras Dayoub, Sean McMahon, Ben Talbot, Ruth Schulz, Peter Corke, Gordon Fraser Wyeth, Ben Upcroft, Michael Milford. 2015
144 citations recorded by the source corpus.
In this paper we focus on the challenging problem of place categorization and semantic mapping on a robot without environment-specific training. Motivated by their ongoing success in various visual recognition tasks, we build our system upon a state-of-the-art convolutional network. We overcome its closed-set limitations by complementing the network with a series of one-vs-all classifiers that can learn to recognize new semantic classes online. Prior domain knowledge is incorporated by embedding the classification system into a Bayesian filter framework that also ensures temporal coherence. We evaluate the classification accuracy of the system on a robot that maps a variety of places on our campus in real-time. We show how semantic information can boost robotic object detection performance and how the semantic map can be used to modulate the robot's behaviour during navigation tasks. The system is made available to the community as a ROS module.
Subfields: Aerospace Engineering, Artificial Intelligence, Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Aerospace Engineering
- Wolfram Burgard
- Niko Sünderhauf
- Dieter Fox
- Michael Milford
- Ben Upcroft
- Sergey Levine
- Peter Corke
- Pieter Abbeel
In Artificial Intelligence
- Sergey Levine
- Dieter Fox
- Pieter Abbeel
- Chelsea Finn
- Wolfram Burgard
- Niko Sünderhauf
- Michael Milford
- Trevor Darrell
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Xiaogang Wang
- Trevor Darrell
- Andrew Zisserman
- Jitendra Malik
- Dacheng Tao
- Andrea Vedaldi
- Bernt Schiele
Citations in the corpus
- The limits and potentials of deep learning for robotics (2018, background): “A combination of place categorization and improved object detection utilizing learned scene-object priors has been demonstrated in [111].”
- Object identification from few examples by improving the invariance of a Deep Convolutional Neural Network (2016, background): “While there are already promising results, robotic vision context comes with its own challenges [6], [7], [8], [9], [10].”
- A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving (2020, background): “[5] 2018 C COCO [122], SceneNet RGB-D [123], QUT Campus Dataset [124] E MC-dropout NMS, Clustering, Sample Statistics Precision and Recall, Absolute Open-set Error, F1-Score”
- Learning and the Unknown: Surveying Steps toward Open World Recognition (2019, background): “The work in this area has received hundreds of citations from a wide range of application areas that need, or are using, OSR including: • Audio analysis (Battaglino, Lepauloux, and Evans 2016; Krstulović 2018), • Automatic Target Recognition (Scherreik and Rigling 2016b; 2016a; Roos and Shaw 2017), ”
- Dropout Sampling for Robust Object Detection in Open-Set Conditions (2017, methodology): “A map was created to convert each COCO class to all corresponding wnids in the dataset.”
- Cross-View Semantic Segmentation for Sensing Surroundings (2019, background): “There is also a huge literature about semantic mapping in robotics domain [1], [2], [8], [9], which provides the semantic abstraction of the environment and a way to communicate with robots.”