TRECVID 2015, 19th International Workshop on Video Retrieval Evaluation, 16-18 November 2015, Gaithersburg, MD, USA
This year EURECOM participated in the TRECVID 2015 Semantic INdexing (SIN) Task  for the submission of four dierent runs for 60 concepts, and Video Hyperlinking (LNK) Task  with the four submissions. Our submission to the SIN Task builds on the runs submitted in the previous years for the 2013 and 2014 SIN tasks, the details of which can be found in  and , while the LNK submissions are based on our previous experiments as in  and . The major changes for 2015 are the use of new Deep Network models to produce extra descriptors for the video shots, and the introduction of various fusion schemes at all levels of the processing, to reduce the problem of overtting. This year, we did not use our uploader model,
partly because of lack of time, and partly because initial experiments showed only marginal improvement after the new features were added. For the LNK Task our approach targeted to connect the textual stream of the videos within the collection and its vocabulary context, as dened by word2vec algorithm, with the output of visual concepts detection tools for the corresponding hyperlinks candidates within one frame-
work. We combined visual concepts detection condence scores with the information about corresponding word vectors distances in order to rerank the baseline text based search. The reranked runs did not outperform the baseline, however they exposed potential of our method for further improvement. Beside this participation, EURECOM took part in the collaborative IRIM submission, the details of this contribution is included in the corresponding publication from the IRIM group. The remainder of this paper brie y describes the descriptors that we have been using, the training and the various fusion schemes, and the content of the submitted runs; and the framework of the condence scores combinations used for reranking in the LNK task.
© NIST. Personal use of this material is permitted. The definitive version of this paper was published in TRECVID 2015, 19th International Workshop on Video Retrieval Evaluation, 16-18 November 2015, Gaithersburg, MD, USA and is available at :