Segmentation and Unsupervised Part-based Discovery of Repetitive Objects

By Rudolph Triebel, Jiwon Shin, Roland Yves Siegwart. 2010

70 citations recorded by the source corpus.

In this paper, we present an unsupervised technique to segment and detect objects in indoor environments. The main idea of this work is to identify object instances whenever there is evidence for at least one other occurence of an object of the same kind. In contrast to former approaches, we do not assume any given segmentation of the data, but instead estimate the segmentation and the existence of object instances concurrently. We apply graph-based clustering in feature and in geometric space to presegmented input data. Each segment is treated as a potential object part, and the inter-dependence of object labels assigned to part clusters are modeled using a Conditional Random Field (CRF) named the “parts graph”. Another CRF is then applied to the scene graph to smooth the class labels using the distributions obtained from the parts graph. First results on indoor 3D laser range data are evaluated and presented.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Bo Li
2. Yoshua Bengio
3. Sergey Levine
4. Li Fei-Fei
5. Andrea Vedaldi
6. Luc Van Gool
7. Trevor Darrell
8. Lei Tai

Citations in the corpus

• Vehicle Detection from 3D Lidar Using Fully Convolutional Network (2016, background): “More delicate segmentation can be obtained by forming graphs on the point cloud [32, 14, 21, 29, 30].”
• On the segmentation of 3D LIDAR point clouds (2011, methodology): “The recent segmentation algorithm of Felzenszwalb and Huttenlocher (FH) [10] for natural images has gained popularity for robotic applications due to its efficiency [13], [15], [16], [19].”
• Real-Time Plane Segmentation Using RGB-D Cameras (2012, methodology): “use Conditional Random Fields to segment and discover repetitive objects in 3D laser range scans [11].”
• Tracking-based semi-supervised learning (2011, background): “Triebel et al. (2010) locate similar objects in dense 3D point clouds.”
• Object discovery in 3D scenes via shape analysis (2013, methodology): “[18], [19], [20] present algorithms for discovering visual categories in laser scans.”
• What could move? Finding cars, pedestrians and bicyclists in 3D laser data (2012, methodology): “However, in contrast to [14], our supervised approach produces object clusters which correspond only to object categories of interest.”