Zero-shot object prediction using semantic scene knowledge
This work addresses the problem of reducing manual labeling efforts for object recognition in zero-shot scenarios, but it is incremental as it builds on existing semantic relation methods.
The paper tackles zero-shot object recognition by using semantic scene knowledge instead of visual attributes, showing that scene knowledge improves object prediction in cluttered scenes where visual recognition is challenging.
This work focuses on the semantic relations between scenes and objects for visual object recognition. Semantic knowledge can be a powerful source of information especially in scenarios with few or no annotated training samples. These scenarios are referred to as zero-shot or few-shot recognition and often build on visual attributes. Here, instead of relying on various visual attributes, a more direct way is pursued: after recognizing the scene that is depicted in an image, semantic relations between scenes and objects are used for predicting the presence of objects in an unsupervised manner. Most importantly, relations between scenes and objects can easily be obtained from external sources such as large scale text corpora from the web and, therefore, do not require tremendous manual labeling efforts. It will be shown that in cluttered scenes, where visual recognition is difficult, scene knowledge is an important cue for predicting objects.