Sergio Guadarrama


YouTube2Text: Recognizing and Describing Arbitrary Activities Using Semantic Hierarchies and Zero-Shot Recognition


Despite a recent push towards large-scale object recognition, activity recognition remains limited to narrow domains and small vocabularies of actions. In this paper, we tackle the challenge of recognizing and describing activities “in-the-wild”. We present a solution that takes a short video clip and outputs a brief sentence that sums up the main activity in the video, such as the actor, the action and its object. Unlike previous work, our approach works on out-of-domain actions: it does not require training videos of the exact activity. If it cannot find an accurate prediction for a pre-trained model, it finds a less specific answer that is also plausible from a pragmatic standpoint. We propose to use semantic hierarchies learned from the data to help to choose an appropriate level of generalization, and pragmatics prior learned from web-scale natural language corpora to penalize unlikely combinations of actors/actions/objects; we also use a web-scale language model to “fill in” novel verbs, i.e. when the verb does not appear in the training set. We evaluate our method on a large YouTube corpus and demonstrate it is able to generate short sentence descriptions of video clips better than baseline approaches.
[PDF] [BibTex]


  • The mapping between the original clip names and our training and test sets can be downloaded from youtube_mapping.txt

  • These are the Subject, Verb, Object SVO Triplets extracted from the descriptions. These are the set of Subjects, Verbs and Objects extracted from the descriptions.

Generating Natural-Language Video Descriptions Using Text-Mined Knowledge


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% of the time.
[PDF] [BibTex]