Fast LSIS Profile Entropy Features for Robot Visual Self-Localization

نویسندگان

  • Hervé Glotin
  • Zhong-Qiu Zhao
  • Emilie Dumont
چکیده

In the Robot Vision task, the participants are asked to answer where is the robot using its vision. The robot may be in 5 rooms (BO-One-person office, CR-Corridor, EO-Two-persons office, KT-Kitchen, PA-Printer Area). In order to train our models we structured the views of each room into several sub-classes: BO-inside, exit, enter; CR-enter, exit, leftstairs, nostairs, rightstairs; EO-enter, exit, inside; KT-enter, exit, cooking hearth, table, television; PA-cabinet, enter, exit. Then an SVM was constructed for each of these 19 sub-classes. After that, we combined the results of SVMs by maximizing to get the final decision. We run our classification models on the new Profile Entropy Features (PEF) that combines RGB color and texture, yielding to one hundred of dimension, and we compare them to generic Descriptor of Fourier (DF). We also made a fusion of the models on these 2 different features. So we got 3 runs. In our experiments, for each decision, we used only the current image, but we do not exploit continuity of the sequences. For this case, a total of 7 teams submitted runs. The official evaluation give for the SVM(PEF) run a score of 544, and for SVM(DF) run a score of -32, while their fusion a score of 509.5. Thus our result possesses the 5th rank over the seven. The experiments show that our SVM model works well with little training cost, and PEF feature works much better than DF feature. It could be concluded that PEF is quite efficient: it is very fast to be computed, with around 10 images computed per second on usual pentium, and less of 2 hours of training (compared to 60 hours for the best systems), but still give a competitive results.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Map-merging in Multi-robot Simultaneous Localization and Mapping Process Using Two Heterogeneous Ground Robots

In this article, a fast and reliable map-merging algorithm is proposed to produce a global two dimensional map of an indoor environment in a multi-robot simultaneous localization and mapping (SLAM) process. In SLAM process, to find its way in this environment, a robot should be able to determine its position relative to a map formed from its observations. To solve this complex problem, simultan...

متن کامل

LSIS TREC VIDEO 2009 High Level Feature Retrieval using Compact Profile Entropy Descriptors

We build a fast video shot retrieval system in the context of the NIST TREC Video 2009 evaluation campaign. We compare our efficient Profile Entropy Features (PEF) to usual features, using various classifiers. These PEF are derived using the projection in the horizontal and vertical orientations. These features are then fed to SVM or KNNG classifiers to produce the keyframe ranks, from which we...

متن کامل

LSIS TREC VIDEO 2008 High Level Feature Shot Segmentation using Compact Profile Entropy and Affinity Propagation Clustering

In this task, we build fast video indexing systems using a kind of efficient features based on the entropy of pixel projections. These features of 45 dimensions, called Profil Entropy Features (PEF), are derived using the projection in the horizontal orientation. These features are then fed to SVMs to produce the keyframe ranks, from which we can get the shot ranks. In the runs, we divided the ...

متن کامل

Self-localization Using Visual Experience Across Domains

— In this study, we aim to solve the single-view robot self-localization problem by using visual experience across domains. Although the bag-of-words method constitutes a popular approach to single-view localization, it fails badly when it's visual vocabulary is learned and tested in different domains. Further, we are interested in using a cross-domain setting, in which the visual vocabulary is...

متن کامل

Vision-based fast and reactive monte-carlo localization

This paper presents a fast approach for vision-based self-localization in RoboCup. The vision system extracts the features required for localization without processing the whole image and is a first step towards independence of lighting conditions. In the field of self-localization, some new ideas are added to the well-known MonteCarlo localization approach that increase both stability and reac...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2009