Recognition and Reconstruction of Transparent Objects for Augmented Reality

Alan Torres Gomez; Walterio W Mayol-Cuevas

doi:10.1109/ISMAR.2014.6948418

Recognition and Reconstruction of Transparent Objects for Augmented Reality

Alan Torres Gomez, Walterio W Mayol-Cuevas

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

7 Citations (Scopus)

408 Downloads (Pure)

Abstract

Dealing with real transparent objects for AR is challenging due to their lack of texture and visual features as well as the drastic changes in appearance as the background, illumination and camera pose change. The few existing methods for glass object detection usually require a carefully controlled environment, specialized illumination hardware or ignore information from different viewpoints.

In this work, we explore the use of a learning approach for classifying transparent objects from multiple images with the aim of both discovering such objects and building a 3D reconstruction to support convincing augmentations. We extract, classify and group small image patches using a fast graph-based segmentation and employ a probabilistic formulation for aggregating spatially consistent glass regions. We demonstrate our approach via analysis of the performance of glass region detection and example 3D reconstructions that allow virtual objects to interact with them.

Original language	English
Title of host publication	2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2014)
Subtitle of host publication	Proceedings of a meeting held 10-12 September 2014, Munich, Germany
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	129-134
Number of pages	6
ISBN (Electronic)	9781479961849
ISBN (Print)	9781479961856
DOIs	https://doi.org/10.1109/ISMAR.2014.6948418
Publication status	Published - Dec 2014

Keywords

Augmented Reality
Glass detection
3D Reconstruction
Computer Vision

Access to Document

10.1109/ISMAR.2014.6948418

Full-text PDF (accepted author manuscript)
This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IEEE at http://ieeexplore.ieee.org/document/6948418/. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 1.83 MB

Handle.net

Persistent link

Professor Walterio W Mayol-Cuevas
- Walterio.Mayol-Cuevasbristol.acuk
- School of Computer Science - Professor in Robotics, Computer Vision and Mobile Systems
- Visual Information Laboratory
Person: Academic , Member

Cite this

Torres Gomez, A., & Mayol-Cuevas, W. W. (2014). Recognition and Reconstruction of Transparent Objects for Augmented Reality. In 2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2014): Proceedings of a meeting held 10-12 September 2014, Munich, Germany (pp. 129-134). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ISMAR.2014.6948418

@inproceedings{c8a6fa86ea65411e9f2b4e784d2a9124,

title = "Recognition and Reconstruction of Transparent Objects for Augmented Reality",

abstract = "Dealing with real transparent objects for AR is challenging due to their lack of texture and visual features as well as the drastic changes in appearance as the background, illumination and camera pose change. The few existing methods for glass object detection usually require a carefully controlled environment, specialized illumination hardware or ignore information from different viewpoints.In this work, we explore the use of a learning approach for classifying transparent objects from multiple images with the aim of both discovering such objects and building a 3D reconstruction to support convincing augmentations. We extract, classify and group small image patches using a fast graph-based segmentation and employ a probabilistic formulation for aggregating spatially consistent glass regions. We demonstrate our approach via analysis of the performance of glass region detection and example 3D reconstructions that allow virtual objects to interact with them.",

keywords = "Augmented Reality, Glass detection, 3D Reconstruction, Computer Vision",

author = "{Torres Gomez}, Alan and Mayol-Cuevas, {Walterio W}",

year = "2014",

month = dec,

doi = "10.1109/ISMAR.2014.6948418",

language = "English",

isbn = "9781479961856",

pages = "129--134",

booktitle = "2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2014)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

}

Torres Gomez, A & Mayol-Cuevas, WW 2014, Recognition and Reconstruction of Transparent Objects for Augmented Reality. in 2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2014): Proceedings of a meeting held 10-12 September 2014, Munich, Germany. Institute of Electrical and Electronics Engineers (IEEE), pp. 129-134. https://doi.org/10.1109/ISMAR.2014.6948418

Recognition and Reconstruction of Transparent Objects for Augmented Reality. / Torres Gomez, Alan; Mayol-Cuevas, Walterio W.
2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2014): Proceedings of a meeting held 10-12 September 2014, Munich, Germany. Institute of Electrical and Electronics Engineers (IEEE), 2014. p. 129-134.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Recognition and Reconstruction of Transparent Objects for Augmented Reality

AU - Torres Gomez, Alan

AU - Mayol-Cuevas, Walterio W

PY - 2014/12

Y1 - 2014/12

N2 - Dealing with real transparent objects for AR is challenging due to their lack of texture and visual features as well as the drastic changes in appearance as the background, illumination and camera pose change. The few existing methods for glass object detection usually require a carefully controlled environment, specialized illumination hardware or ignore information from different viewpoints.In this work, we explore the use of a learning approach for classifying transparent objects from multiple images with the aim of both discovering such objects and building a 3D reconstruction to support convincing augmentations. We extract, classify and group small image patches using a fast graph-based segmentation and employ a probabilistic formulation for aggregating spatially consistent glass regions. We demonstrate our approach via analysis of the performance of glass region detection and example 3D reconstructions that allow virtual objects to interact with them.

AB - Dealing with real transparent objects for AR is challenging due to their lack of texture and visual features as well as the drastic changes in appearance as the background, illumination and camera pose change. The few existing methods for glass object detection usually require a carefully controlled environment, specialized illumination hardware or ignore information from different viewpoints.In this work, we explore the use of a learning approach for classifying transparent objects from multiple images with the aim of both discovering such objects and building a 3D reconstruction to support convincing augmentations. We extract, classify and group small image patches using a fast graph-based segmentation and employ a probabilistic formulation for aggregating spatially consistent glass regions. We demonstrate our approach via analysis of the performance of glass region detection and example 3D reconstructions that allow virtual objects to interact with them.

KW - Augmented Reality

KW - Glass detection

KW - 3D Reconstruction

KW - Computer Vision

U2 - 10.1109/ISMAR.2014.6948418

DO - 10.1109/ISMAR.2014.6948418

M3 - Conference Contribution (Conference Proceeding)

SN - 9781479961856

SP - 129

EP - 134

BT - 2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2014)

PB - Institute of Electrical and Electronics Engineers (IEEE)

ER -

Torres Gomez A, Mayol-Cuevas WW. Recognition and Reconstruction of Transparent Objects for Augmented Reality. In 2014 IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2014): Proceedings of a meeting held 10-12 September 2014, Munich, Germany. Institute of Electrical and Electronics Engineers (IEEE). 2014. p. 129-134 Epub 2014 Nov 6. doi: 10.1109/ISMAR.2014.6948418

Recognition and Reconstruction of Transparent Objects for Augmented Reality

Abstract

Keywords

Access to Document

Handle.net

Fingerprint

Profiles

Professor Walterio W Mayol-Cuevas

Cite this