In this paper we present a novel technique for object modeling and object recognition in video. Given a setof videos containing 360 degrees views of objects we compute a model for each object, then we analyzeshort videos to determine if the object depicted in the video is one of the modeled objects. The object modelis built from a video spanning a 360 degree view of the object taken against a uniform background. In orderto create the object model, the proposed techniques selects a few representative frames from each video andlocal features of such frames. The object recognition is performed selecting a few frames from the queryvideo, extracting local features from each frame and looking for matches in all the representative framesconstituting the models of all the objects. If the number of matches exceed a fixed threshold thecorresponding object is considered the recognized objects .To evaluate our approach we acquired a datasetof 25 videos representing 25 different objects and used these videos to build the objects model. Then wetook 25 test videos containing only one of the known objects and 5 videos containing only unknown objects.Experiments showed that, despite a significant compression in the model, recognition results aresatisfactory.
|Number of pages||9|
|Publication status||Published - 2014|
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition