University of Cambridge > Talks.cam > Inference Group > LOCUS: Learning Object Classes with Unsupervised Segmentation

LOCUS: Learning Object Classes with Unsupervised Segmentation

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Phil Cowans.

LOCUS (Learning Object Classes with Unsupervised Segmentation) is a system for learning object class models and object segmentations from unannotated images. LOCUS uses a generative probabilistic model to combine bottom-up cues of color and edge with top-down cues of shape and pose. A key aspect of the model is that the object appearance is allowed to vary from image to image, allowing for significant within-class variation.

I will show that LOCUS successfully learns class models from unlabelled images, whilst also giving segmentation accuracies that rival existing supervised segmentation methods. LOCUS also infers the position and pose of the object in each image.

The intention is to use these class models for simultaneous recognition and segmentation of objects. I will present some preliminary results for this task and discuss some promising extensions to the LOCUS model. Finally, I will demonstrate that LOCUS can be used to perform motion segmentation and object tracking in video, despite changes in illumination, pose and background clutter.

This is joint work with Nebojsa Jojic of MSR Redmond

This talk is part of the Inference Group series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2019 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity