We present a holistic data-driven technique that generates natural-language descriptions for videos. We combine the output of state-of-the-art object and activity detectors with "real-world" knowledge to select the most probable subject-verb-object triplet for describing a video. We show that this knowledge, automatically mined from web-scale text corpora, enhances the triplet selection algorithm by providing it contextual information and leads to a four-fold increase in activity identification. Unlike previous methods, our approach can annotate arbitrary videos without requiring the expensive collection and annotation of a similar training video corpus. We evaluate our technique against a baseline that does not use text-mined knowledge and show that humans prefer our descriptions 61 percent of the time.

PDF

In Proceedings of the 27th AAAI Conference on Artificial Intelligence (AAAI-2013), pp. 541--547, July 2013.

@inproceedings{krishnamoorthy:aaai13, title={Generating Natural-Language Video Descriptions Using Text-Mined Knowledge}, author={Niveda Krishnamoorthy and Girish Malkarnenkar and Raymond J. Mooney and Kate Saenko and Sergio Guadarrama}, booktitle={Proceedings of the 27th AAAI Conference on Artificial Intelligence (AAAI-2013)}, month={July}, pages={541--547}, url="http://www.cs.utexas.edu/users/ai-lab?krishnamoorthy:aaai13", year={2013} }

Slides (PPT)