Skip to Main content Skip to Navigation
Conference papers

Reconstruction of transparent objects in unstructured scenes with a depth camera

Nicolas Alt 1 Patrick Rives 2 Eckehard Steinbach 1
2 Lagadic - Visual servoing in robotics, computer vision, and augmented reality
CRISAM - Inria Sophia Antipolis - Méditerranée , Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : The visual 3D reconstruction of transparent objects in unstructured scenes is challenging due to the complex image formation principles underlying their visual appearance. Most state-of-the-art reconstruc- tion methods ignore this problem and assume Lambertian reflection. Yet, transparent objects are relevant scene information for applica- tions in intelligent robotics (such as grasping) or virtual reality. In this work, we present an approach to detect non-planar transparent objects, like bottles or glasses, by specifically searching for geom- etry inconsistencies caused by refraction or reflection. Depth infor- mation is acquired using a Kinect sensor, which is moved within the scene in order to acquire multiple views. The individual measure- ments are combined into a 3D volume, yielding the objects' loca- tion and a rough shape estimate. Results are presented using various household objects made of glass or plastic.
Document type :
Conference papers
Complete list of metadata

Cited literature [10 references]  Display  Hide  Download

https://hal.inria.fr/hal-00845456
Contributor : Eric Marchand <>
Submitted on : Wednesday, July 17, 2013 - 10:34:30 AM
Last modification on : Wednesday, June 16, 2021 - 3:41:57 AM
Long-term archiving on: : Friday, October 18, 2013 - 4:22:57 AM

File

2013_icip_alt.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00845456, version 1

Citation

Nicolas Alt, Patrick Rives, Eckehard Steinbach. Reconstruction of transparent objects in unstructured scenes with a depth camera. IEEE Int. Conf. on Image Processing, ICIP'13, Sep 2013, Melbourne, Australia. ⟨hal-00845456⟩

Share

Metrics

Record views

943

Files downloads

885