What is (Missing or Wrong) in the Scene? A Hybrid Deep Boltzmann Machine for Contextualized Scene Modeling

Download

index.pdf

Date

2018-05-25

Author

Bozcan, Ilker
Oymak, Yağmur
Alemdar, İdil Zeynep
Kalkan, Sinan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

82
views

38
downloads

Scene models allow robots to reason about what is in the scene, what else should be in it, and what should not be in it. In this paper, we propose a hybrid Boltzmann Machine (BM) for scene modeling where relations between objects are integrated. To be able to do that, we extend BM to include tri-way edges between visible (object) nodes and make the network to share the relations across different objects. We evaluate our method against several baseline models (Deep Boltzmann Machines, and Restricted Boltzmann Machines) on a scene classification dataset, and show that it performs better in several scene reasoning tasks.

Subject Keywords

Training , Task Analysis , Robots , Computational Modeling , Context Modeling , Estimation , Cognition

URI

https://hdl.handle.net/11511/47738

DOI

https://doi.org/10.1109/icra.2018.8460828

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

COSMO: Contextualized scene modeling with Boltzmann Machines Bozcan, Ilker; Kalkan, Sinan (Elsevier BV, 2019-03-01) Scene modeling is very crucial for robots that need to perceive, reason about and manipulate the objects in their environments. In this paper, we adapt and extend Boltzmann Machines (BMs) for contextualized scene modeling. Although there are many models on the subject, ours is the first to bring together objects, relations, and affordances in a highly-capable generative model. For this end, we introduce a hybrid version of BMs where relations and affordances are incorporated with shared, tri-way connections...
A Deep Incremental Boltzmann Machine for Modeling Context in Robots Doğan, Fethiye Irmak; Çelikkanat, Hande; Kalkan, Sinan (2018-05-25) Context is an essential capability for robots that are to be as adaptive as possible in challenging environments. Although there are many context modeling efforts, they assume a fixed structure and number of contexts. In this paper, we propose an incremental deep model that extends Restricted Boltzmann Machines. Our model gets one scene at a time, and gradually extends the contextual model when necessary, either by adding a new context or a new context layer to form a hierarchy. We show on a scene classific...
Contextualized scene modeling using boltzmann machines Bozcan, İlker; Kalkan, Sinan; Department of Computer Engineering (2018) Scene modeling is very crucial for robots that need to perceive, reason about and manipulate the objects in their environments. In this thesis, we propose a variant of Boltzmann Machines (BMs) for contextualized scene modeling. Although many computational models have been proposed for the problem, ours is the first to bring together objects, relations, and affordances in a highly-capable generative model. For this end, we introduce a hybrid version of BMs where relations and affordances are introduced with ...
CINet: A Learning Based Approach to Incremental Context Modeling in Robots Doğan, Fethiye Irmak; Bozcan, Ilker; Çelik, Mehmet; Kalkan, Sinan (2018-10-05) There have been several attempts at modeling context in robots. However, either these attempts assume a fixed number of contexts or use a rule-based approach to determine when to increment the number of contexts. In this paper, we pose the task of when to increment as a learning problem, which we solve using a Recurrent Neural Network. We show that the network successfully (with 98% testing accuracy) learns to predict when to increment, and demonstrate, in a scene modeling problem (where the correct number ...
Optimal control of a half circular compliant legged monopod Özkan Aydın, Yasemin; Leblebicioğlu, Mehmet Kemal; Saranlı, Afşar; Department of Electrical and Electronics Engineering (2013) Legged robots have complex architecture because of their nonlinear dynamics and unpredictable ground contact characteristics. They can be also dynamically stable and exhibit dynamically dexterous behaviors like running, jumping, flipping which require complex plant models that may sometimes be difficult to build. In this thesis, we focused on half circular compliant legged monopod that can be considered as a reduced-order dynamical model for the hexapod robot, called RHex. The main objective of this thesis ...

Citation Formats

I. Bozcan, Y. Oymak, İ. Z. Alemdar, and S. Kalkan, “What is (Missing or Wrong) in the Scene? A Hybrid Deep Boltzmann Machine for Contextualized Scene Modeling,” 2018, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/47738.