Readspeaker menu

Urban Scene Understanding

Where exactly are the things that usually occur in urban scenes in my image?

Sven Sickert, Marcel Simon, Erik Rodner, and Joachim Denzler


Incorporating Spatial Priors in Convolutional Networks

 

[Brust15:CPN]

Clemens-Alexander Brust and Sven Sickert and Marcel Simon and Erik Rodner and Joachim Denzler. Convolutional Patch Networks with Spatial Prior for Road Detection and Urban Scene Understanding . International Conference on Computer Vision Theory and Applications (VISAPP). 510-517. 2015. [bib]

[Brust15:ECP]

Clemens-Alexander Brust and Sven Sickert and Marcel Simon and Erik Rodner and Joachim Denzler. Efficient Convolutional Patch Networks for Scene Understanding . CVPR Workshop on Scene Understandind (CVPR-WS). 2015. [bib]
Urban Scene Understanding Teaser
 
  Classifying single image patches is important in many different applications, such as road detection or scene understanding. In this paper, we present convolutional patch networks, which are convolutional networks learned to distinguish different image patches and which can be used for pixel-wise labeling. We also show how to incorporate spatial information of the patch as an input to the network, which allows for learning spatial priors for certain categories jointly with an appearance model. In particular, we focus on road detection and urban scene understanding, two application areas where we are able to achieve state-of-the-art results on the KITTI as well as on the LabelMeFacade dataset. Furthermore, our paper offers a guideline for people working in the area and desperately wandering through all the painstaking details that render training CNs on image patches extremely difficult.  
 

Code is available on GitHub! If you want to know more, please visit our GitHub Page!

The dataset is also available on GitHub. More information can be found in our section on datasets.


Exploitation of Context Cues using Iterative Context Forests

 

[Froehlich12:SSM]

Björn Fröhlich and Erik Rodner and Joachim Denzler. Semantic Segmentation with Millions of Features: Integrating Multiple Cues in a Combined Random Forest Approach. Asian Conference on Computer Vision (ACCV).2012. [bib]

[Froehlich12:ATG]

Björn Fröhlich and Erik Rodner and Joachim Denzler. As Time Goes By - Anytime Semantic Segmentation with Iterative Context Forests. Symposium of the German Association for Pattern Recognition (DAGM).2012. [bib]
SemSegTeaser.png
 
  Semantic interpretation and understanding of images is an important goal of visual recognition research and offers a large variety of possible applications. One step towards this goal is semantic segmentation, which aims for automatic labeling of image regions and pixels
with category names. Since usual images contain several millions of pixel, the use of kernel-based methods for the task of semantic segmentation is limited due to the involved computation times. In this paper, we overcome this drawback by exploiting efficient kernel calculations using the histogram intersection kernel for fast and exact Gaussian process classification. Our results show that non-parametric Bayesian methods can be utilized for semantic segmentation without sparse approximation techniques. Furthermore, in experiments, we show a significant benefit in terms of classification accuracy compared to state-of-the-art methods.
 


Large-scale Gaussian Process Inference for Semantic Segmentation

 

[Freytag12:ESS]

Alexander Freytag and Björn Fröhlich and Erik Rodner and Joachim Denzler. Efficient Semantic Segmentation with Gaussian Processes and Histogram Intersection Kernels. International Conference on Pattern Recognition (ICPR).2012. 3313--3316. (Oral). [project page][bib]
hikSemSeg.png
  Semantic interpretation and understanding of images is an important goal of visual recognition research and offers a large variety of possible applications. One step towards this goal is semantic segmentation, which aims for automatic labeling of image regions and pixels
with category names. Since usual images contain several millions of pixel, the use of kernel-based methods for the task of semantic segmentation is limited due to the involved computation times. In this paper, we overcome this drawback by exploiting efficient kernel calculations using the histogram intersection kernel for fast and exact Gaussian process classification. Our results show that non-parametric Bayesian methods can be utilized for semantic segmentation without sparse approximation techniques. Furthermore, in experiments, we show a significant benefit in terms of classification accuracy compared to state-of-the-art methods.