Open AccessProceedings Article10.1109/IJCNN.2012.6252606

Developmental approach for interactive object discovery

- 10 Jun 2012

- pp 1-7

TL;DR: A visual system for a humanoid robot that supports an efficient online learning and recognition of various elements of the environment and is implemented on the iCub robot and shows 85% average recognition rate for 10 objects after 30 minutes of interaction.

Abstract: We present a visual system for a humanoid robot that supports an efficient online learning and recognition of various elements of the environment Taking inspiration from child's perception and following the principles of developmental robotics, our algorithm does not require image databases, predefined objects nor face/skin detectors The robot explores the visual space from interactions with people and its own experiments The object detection is based on the hypothesis of coherent motion and appearance during manipulations A hierarchical object representation is constructed from SURF points and color of superpixels that are grouped in local geometric structures and form the basis of a multiple-view object model The learning algorithm accumulates the statistics of feature occurrences and identifies objects using a maximum likelihood approach and temporal coherency The proposed visual system is implemented on the iCub robot and shows 85% average recognition rate for 10 objects after 30 minutes of interaction

AI Agents for this Paper

Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps

Most frequently asked questions

1. What future works have the authors mentioned in the paper "Developmental approach for interactive object discovery" ?

In future work, this system will be used as the basis for object categorization through robot experiments and affordance learning.. The authors are planning to use the mutual information between the data from robot ’ s motors and the behavior of moving regions in the visual space, by analogy with [ 27 ].. The kinematic model can be used to improve categorization by predicting the position of robot ’ s parts.

2. What are the contributions mentioned in the paper "Developmental approach for interactive object discovery" ?

The authors present a visual system for a humanoid robot that supports an efficient online learning and recognition of various elements of the environment.

Fig. 1. Proto-object segmentation and intermediate results of the image processing.

Fig. 6. Experimental object O1 −O10 from top-left to bottom-right.

Fig. 4. Multi-view object recognition: object views from different perspectives are recognized as the same object.

Fig. 5. Multiple objects recognition: an object and human hand, composing a single moving blob, are recognized as separate objects.

Fig. 3. View recognition through the voting method.

Citations

•Journal Article•10.1109/TAMD.2013.2280614

Object Learning Through Active Exploration

Serena Ivaldi, +7 more

- 01 Mar 2014

- IEEE Transactions on Autonomous Mental D...

TL;DR: This paper presents the design guidelines of the cognitive architecture, its main functionalities, and outlines the cognitive process of the robot by showing how it learns to recognize objects in a human-robot interaction scenario inspired by social parenting.

...read moreread less

•Journal Article•10.1016/J.ROBOT.2014.11.005

Deep unsupervised network for multimodal perception, representation and classification

Alain Droniou, +2 more

- 01 Sep 2015

- Robotics and Autonomous Systems

TL;DR: An architecture based on deep networks, which is used by the humanoid robot iCub to learn a task from multiple perceptual modalities is proposed, which performs a substantial dimensionality reduction by providing both a symbolic representation of data and a fine discrimination between two similar stimuli.

...read moreread less

•Proceedings Article•10.1109/DEVLRN.2013.6652525

Learning to recognize objects through curiosity-driven manipulation with the iCub humanoid robot

Sao Mai Nguyen, +8 more

- 04 Nov 2013

TL;DR: Experimental results show that the humanoid robot iCub can learn to recognize 3D objects by manipulation and in interaction with teachers by choosing the adequate exploration strategy to enhance competence progress and by focusing its efforts on the most complex tasks.

...read moreread less

•Journal Article•10.1371/JOURNAL.PONE.0140732

MCA-NMF: Multimodal Concept Acquisition with Non-Negative Matrix Factorization.

Olivier Mangin, +6 more

- 21 Oct 2015

- PLOS ONE

TL;DR: The model finds patterns in multimodal sensor input that characterize associations across modalities (speech utterances, images and motion) and proposes this computational model as an answer to the question of how some class of concepts can be learnt.

...read moreread less

•Proceedings Article•10.1109/HUMANOIDS.2012.6651528

Perception and human interaction for developmental learning of objects and affordances

Serena Ivaldi, +7 more

- 29 Nov 2012

TL;DR: A cognitive architecture for humanoids interacting with objects and caregivers in a developmental robotics scenario, focusing on the elementary action, perception and interaction modules is described.

...read moreread less

...

Expand

References

Journal Article•10.1023/B:VISI.0000029664.99615.94

Distinctive Image Features from Scale-Invariant Keypoints

David G. Lowe

- 01 Nov 2004

- International Journal of Computer Vision

TL;DR: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene and can robustly identify objects among clutter and occlusion while achieving near real-time performance.

...read moreread less

59.3K

•Journal Article•10.1016/J.CVIU.2007.09.014

Speeded-Up Robust Features (SURF)

Herbert Bay, +3 more

- 01 Jun 2008

- Computer Vision and Image Understanding

TL;DR: A novel scale- and rotation-invariant detector and descriptor, coined SURF (Speeded-Up Robust Features), which approximates or even outperforms previously proposed schemes with respect to repeatability, distinctiveness, and robustness, yet can be computed and compared much faster.

...read moreread less

14.9K

Journal Article•10.1023/B:VISI.0000013087.49260.FB

Robust Real-Time Face Detection

Paul A. Viola, +1 more

- 01 May 2004

- International Journal of Computer Vision

TL;DR: In this paper, a face detection framework that is capable of processing images extremely rapidly while achieving high detection rates is described. But the detection performance is limited to 15 frames per second.

...read moreread less

14.6K

Proceedings Article•10.1109/ICCV.2001.937709

Robust real-time face detection

Paul A. Viola, +1 more

- 07 Jul 2001

TL;DR: A new image representation called the “Integral Image” is introduced which allows the features used by the detector to be computed very quickly and a method for combining classifiers in a “cascade” which allows background regions of the image to be quickly discarded while spending more computation on promising face-like regions.

...read moreread less

13.8K

Proceedings Article•10.1109/CVPR.1994.323794

Good features to track

Jianbo Shi, +1 more

- 21 Jun 1994

TL;DR: A feature selection criterion that is optimal by construction because it is based on how the tracker works, and a feature monitoring method that can detect occlusions, disocclusions, and features that do not correspond to points in the world are proposed.

...read moreread less

...

Expand

Developmental approach for interactive object discovery

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What future works have the authors mentioned in the paper "Developmental approach for interactive object discovery" ?

2. What are the contributions mentioned in the paper "Developmental approach for interactive object discovery" ?

Figures

Citations

Object Learning Through Active Exploration

Deep unsupervised network for multimodal perception, representation and classification

Learning to recognize objects through curiosity-driven manipulation with the iCub humanoid robot

MCA-NMF: Multimodal Concept Acquisition with Non-Negative Matrix Factorization.

Perception and human interaction for developmental learning of objects and affordances

References

Distinctive Image Features from Scale-Invariant Keypoints

Speeded-Up Robust Features (SURF)

Robust Real-Time Face Detection

Robust real-time face detection

Good features to track

Related Papers (5)

Receptive field cooccurrence histograms for object detection

Autonomous acquisition of generic handheld objects in unstructured environments via sequential back-tracking for object recognition

An interactive open-ended learning approach for 3D object recognition

Unsupervised discovery of object classes with a mobile robot

Move and the robot will learn: Vision-based autonomous learning of object models