Fine-grained semi-supervised labeling of large shape collections
- 1 November 2013
- journal article
- research article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 32 (6), 1-10
- https://doi.org/10.1145/2508363.2508364
Abstract
In this paper we consider the problem of classifying shapes within a given category (e.g., chairs) into finer-grained classes (e.g., chairs with arms, rocking chairs, swivel chairs). We introduce a multi-label (i.e., shapes can belong to multiple classes) semi-supervised approach that takes as input a large shape collection of a given category with associated sparse and noisy labels, and outputs cleaned and complete labels for each shape. The key idea of the proposed approach is to jointly learn a distance metric for each class which captures the underlying geometric similarity within that class, e.g., the distance metric for swivel chairs evaluates the global geometric resemblance of chair bases. We show how to achieve this objective by first geometrically aligning the input shapes, and then learning the class-specific distance metrics by exploiting the feature consistency provided by this alignment. The learning objectives consider both labeled data and the mutual relations between the distance metrics. Given the learned metrics, we apply a graph-based semi-supervised classification technique to generate the final classification results. In order to evaluate the performance of our approach, we have created a benchmark data set where each shape is provided with a set of ground truth labels generated by Amazon's Mechanical Turk users. The benchmark contains a rich variety of shapes in a number of categories. Experimental results show that despite this variety, given very sparse and noisy initial labels, the new method yields results that are superior to state-of-the-art semi-supervised learning techniques.Keywords
Funding Information
- Air Force Office of Scientific Research (FA9550-12-1-0372)
- Office of Naval Research (N00014-13-1-0341)
- National Science Foundation (FODAVA 808515, CCF 1011228)
- Max Planck Center for Visual Computing and Communications
- Division of Computing and Communication Foundations (FODAVA 808515, CCF 1011228)
This publication has 25 references indexed in Scilit:
- Style-content separation by anisotropic part scalesACM Transactions on Graphics, 2010
- Semi-supervised distance metric learning for collaborative image retrieval and clusteringACM Transactions on Multimedia Computing, Communications, and Applications, 2010
- Distributed Optimization and Statistical Learning via the Alternating Direction Method of MultipliersFoundations and Trends® in Machine Learning, 2010
- Exact Matrix Completion via Convex OptimizationFoundations of Computational Mathematics, 2009
- Global Correspondence Optimization for Non‐Rigid Registration of Depth ScansComputer Graphics Forum, 2008
- On Visual Similarity Based 3D Model RetrievalComputer Graphics Forum, 2003
- Normalized cuts and image segmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- Using spin images for efficient object recognition in cluttered 3D scenesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1999
- WordNetCommunications of the ACM, 1995
- Robust regression using iteratively reweighted least-squaresCommunications in Statistics - Theory and Methods, 1977