Abstract | ||
---|---|---|
Motivated by the problem of learning to detect and recognize objects with minimal supervision, we develop a hierarchical probabilistic model for the spatial structure of visual scenes. In contrast with most existing models, our approach explicitly captures uncertainty in the number of object instances depicted in a given image. Our scene model is based on the transformed Dirichlet process (TDP), a novel extension of the hier- archical DP in which a set of stochastically transformed mixture com- ponents are shared between multiple groups of data. For visual scenes, mixture components describe the spatial structure of visual features in an object-centered coordinate frame, while transformations model the ob- ject positions in a particular image. Learning and inferenc e in the TDP, which has many potential applications beyond computer vision, is based on an empirically effective Gibbs sampler. Applied to a dataset of par- tially labeled street scenes, we show that the TDP's inclusi on of spatial structure improves detection performance, flexibly exploi ting partially labeled training images. |
Year | Venue | Keywords |
---|---|---|
2005 | NIPS | probabilistic model,gibbs sampler,computer vision |
Field | DocType | Citations |
Dirichlet process,Pattern recognition,Minimal supervision,Inference,Computer science,Artificial intelligence,Statistical model,Dirichlet distribution,Spatial structure,Machine learning,Gibbs sampling | Conference | 55 |
PageRank | References | Authors |
10.33 | 16 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Erik B. Sudderth | 1 | 1420 | 119.04 |
Antonio Torralba | 2 | 14607 | 956.27 |
William T. Freeman | 3 | 17382 | 1968.76 |
Alan S. Willsky | 4 | 7466 | 847.01 |