Object Co-detection

Sid Yingze Bao, Yu Xiang, and Silvio Savarese
Computer Vision Lab, University of Michigan at Ann Arbor

What is it about?
We introduce a new problem object co-detection. Given a set of images with objects observed from two or multiple images, the goal of co-detection is to detect the objects, establish the identity of individual object instance, as well as estimate the viewpoint transformation of corresponding object instances. In designing a co-detector, we follow the intuition that an object has consistent appearance when observed from the same or different viewpoints. By modeling an object using state-of-the-art part-based representations, we measure appearance consistency between objects by comparing part appearance and geometry across images. This allows to effectively account for object self-occlusions and viewpoint transformations. Extensive experimental evaluation indicates that our co-detector obtains more accurate detection results than if objects were to be detected from each image individually. Moreover, we demonstrate the relevance of our co-detection scheme to other recognition problems such as single instance object recognition, wide-baseline matching, and image query.

single image detection result co-detection result
(a) Single image object detection. Notice miss positives and false alarms.
(b) Object co-detection. Different colors correspond to different matching objects. Co-detection recovers missed positives and removes false alarms, compared to single image object detection.

May 2013
  • We posted the image lists for our experimental evaluation. We would like to hear your result!

Papers and

  • Sid Yingze Bao, Yu Xiang, and Silvio Savarese, Object Co-detection, Proceedings of the European Conference on Computer Vision, 2012. [pdf] [bibtex]

  • Data sets
    We evaluated our algorithm using image pairs extracted from the following datasets.

    Last update: 5/15/13