Brno University of Technology at TRECVid 2009

Brno University of Technology at TRECVid 2009

Paper in proceedings outside WoS and Scopus

Inthis paper we describe our experiments in High-level featureextraction (HLF) and Search tasks of the 2009 TRECVid evaluation.This year, we have concentrated mainly on the local (affinecovariant) image features and their transformation into a searchableform, especially using the indexing techniques.

In brief, we have submitted thefollowing runs:

1. HLF: We have used training method based on support vector machine (SVM) using five types of global and local image features. Results were submitted in the BRNO_HLF_SI run.

2. Search: We have performed a fully automatic experiment based on the transformed local image features together with face detection and global features - color layout and texture features in the BrnoUT_visual.2 run.

The paper is organized asfollows. In Section 1, a motivation and an overview of the work ispresented. We dedicated Section 2 to the feature extraction task,which is being used in common by the HLF and Search tasks. Details ofthe tasks we have sent are given in Section 3 and Section 4. Finally,Section 5 discusses the achieved results and concludes the paper. Theresult tables are attached at the end of the paper.

Inthis paper we describe our experiments in High-level featureextraction (HLF) and Search tasks of the 2009 TRECVid evaluation.This year, we have concentrated mainly on the local (affinecovariant) image features and their transformation into a searchableform, especially using the indexing techniques.

In brief, we have submitted thefollowing runs:

1. HLF: We have used training method based on support vector machine (SVM) using five types of global and local image features. Results were submitted in the BRNO_HLF_SI run.

2. Search: We have performed a fully automatic experiment based on the transformed local image features together with face detection and global features - color layout and texture features in the BrnoUT_visual.2 run.

The paper is organized asfollows. In Section 1, a motivation and an overview of the work ispresented. We dedicated Section 2 to the feature extraction task,which is being used in common by the HLF and Search tasks. Details ofthe tasks we have sent are given in Section 3 and Section 4. Finally,Section 5 discusses the achieved results and concludes the paper. Theresult tables are attached at the end of the paper.

TRECVid, Search, HLF, NIST

TRECVid, Search, HLF, NIST

CHMELAŘ, P.; BERAN, V.; HEROUT, A.; HRADIŠ, M.; ŘEZNÍČEK, I.; ZEMČÍK, P.

16.11.2009

National Institute of Standards and Technology

Gaithersburg, MD

TRECVID 2009: Participant Notebook Papers and Slides

1

11

11