Automatic Semantic Video Annotation

Automatic Semantic Video Annotation

Amjad Altadmri, Amr Ahmed*, Andrew Hunter

Poster - Click here to download PDF.
Poster – see link below to download PDF.

 

 

 

(Click Semantic Video Annotation-with Knowledge ” http://amrahmed.blogs.lincoln.ac.uk/files/2013/03/Semantic-Video-Annotation-with-Knowledge.pdf  ,  to download the pdf)

INTRODUCTION

The volume of video data is growing exponentially. This data need to be annotated to facilitate search and retrieval, so that we can quickly find a video whenever needed.

Manual Annotation, especially for such volume, is time consuming and would be expensive. Hence, automated annotation systems are required.

 

AIM

Automated Semantic Annotation of wide-domain videos (i.e. no domain restrictions). This is an important step towards bridging the “Semantic Gap” in video understanding.

 

METHOD

1. Extracting “Video Signature” for each  video.
2. Match signatures to find most similar  videos, with annotations
3. Analyse and process obtained annotations, in consultation with Common-sense knowledge-bases
4. Produce the suggested annotation.

EVALUATION

• Two standard, and challenging  Datasets  were used. TRECVID BBC Rush and UCF.
• Black-box and White-box testing carried out.
•Measures include: Precision, Confusion Matrix.

CONCLUSION

•Developed an Automatic Semantic Video Annotation framework.
•Not restricted to a specific domain videos.
•Utilising Common-sense Knowledge enhances scene understanding and improve semantic annotation.
Publications
  1. A framework for automatic semantic video annotation 
    Altadmri, Amjad and Ahmed, Amr (2013) A framework for automatic semantic video annotation. Multimedia Tools and Applications, 64 (2). ISSN 1380-7501.
  2. Semantic levels of domain-independent commonsense knowledgebase for visual indexing and retrieval applications 
    Altadmri, Amjad and Ahmed, Amr and Mohtasseb Billah, Haytham (2012) Semantic levels of domain-independent commonsense knowledgebase for visual indexing and retrieval applications. Neural Information Processing. Lecture Notes in Computer Science, 7663 . pp. 640-647. ISSN 0302-9743
  3. VisualNet: commonsense knowledgebase for video and image indexing and retrieval application 
    Alabdullah Altadmri, Amjad and Ahmed, Amr (2009) VisualNet: commonsense knowledgebase for video and image indexing and retrieval application. In: IEEE International Conference on Intelligent Computing and Intelligent Systems, 21-22 November 2009, Shanghai, China..
  4. Automatic semantic video annotation in wide domain videos based on similarity and commonsense knowledgebases 
    Altadmri, Amjad and Ahmed, Amr (2009) Automatic semantic video annotation in wide domain videos based on similarity and commonsense knowledgebases. In: The IEEE International Conference on Signal and Image Processing Applications (ICSIPA 2009), 18-19th November 2009, Malaysia.
  5. Video databases annotation enhancing using commonsense knowledgebases for indexing and retrieval 
    Altadmri, Amjad and Ahmed, Amr (2009) Video databases annotation enhancing using commonsense knowledgebases for indexing and retrieval. In: The 13th IASTED International Conference on Artificial Intelligence and Soft Computing., September 7 � 9, 2009, Palma de Mallorca, Spain.

Amjad Altadmri – PhD

Amjad Altadmri has passed his PhD viva, subject to minor amendments, earlier today.

Thesis Title:  “Semantic Video Annotation in Domain-Independent Videos Utilising Similarity and Commonsense Knowledgebases

Thanks to the external, Dr John Wood from the University of Essex, the internal Dr Bashir Al-Diri and the viva chair, Dr Kun Guo.

Congratulations and Well done.

All colleagues are invited to join Amjad on celebrating his achievement, tomorrow (Thursday 28th Feb) at 12:00noon, in our meeting room MC3108, with some drinks and light refreshments available.

Best wishes.

New Journal paper Accepted to the “Multimedia Tools and Applications”

New Journal paper accepted for publishing in the Journal of “Multimedia Tools and Applications“.

The paper title is “A Framework for Automatic Semantic Video Annotation utilising Similarity and Commonsense Knowledgebases

Abstract:

The rapidly increasing quantity of publicly available videos has driven research into developing automatic tools for indexing, rating, searching and retrieval. Textual semantic representations, such as tagging, labelling and annotation, are often important factors in the process of indexing any video, because of their user-friendly way of representing the semantics appropriate for search and retrieval. Ideally, this annotation should be inspired by the human cognitive way of perceiving and of describing videos. The difference between the low-level visual contents and the corresponding human perception is referred to as the ‘semantic gap’. Tackling this gap is even harder in the case of unconstrained videos, mainly due to the lack of any previous information about the analyzed video on the one hand, and the huge amount of generic knowledge required on the other.

This paper introduces a framework for the Automatic Semantic Annotation of unconstrained videos. The proposed framework utilizes two non-domain-specific layers: low-level visual similarity matching, and an annotation analysis that employs commonsense knowledgebases. Commonsense ontology is created by incorporating multiple-structured semantic relationships. Experiments and black-box tests are carried out on standard video databases for
action recognition and video information retrieval. White-box tests examine the performance of the individual intermediate layers of the framework, and the evaluation of the results and the statistical analysis show that integrating visual similarity matching with commonsense semantic relationships provides an effective approach to automated video annotation.

 

Well done and congratulations to Amjad Altadmri .