Back to Search
Start Over
Large-scale knowledge transfer for object localization in ImageNet
- Source :
- CVPR, Guillaumin, M & Ferrari, V 2012, Large-scale knowledge transfer for object localization in ImageNet . in Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on . pp. 3202-3209 . https://doi.org/10.1109/CVPR.2012.6248055
- Publication Year :
- 2012
- Publisher :
- IEEE, 2012.
-
Abstract
- ImageNet is a large-scale database of object classes with millions of images. Unfortunately only a small fraction of them is manually annotated with bounding-boxes. This prevents useful developments, such as learning reliable object detectors for thousands of classes. In this paper we propose to automatically populate ImageNet with many more bounding-boxes, by leveraging existing manual annotations. The key idea is to localize objects of a target class for which annotations are not available, by transferring knowledge from related source classes with available annotations. We distinguish two kinds of source classes: ancestors and siblings. Each source provides knowledge about the plausible location, appearance and context of the target objects, which induces a probability distribution over windows in images of the target class. We learn to combine these distributions so as to maximize the location accuracy of the most probable window. Finally, we employ the combined distribution in a procedure to jointly localize objects in all images of the target class. Through experiments on 0.5 million images from 219 classes we show that our technique (i) annotates a wide range of classes with bounding-boxes; (ii) effectively exploits the hierarchical structure of ImageNet, since all sources and types of knowledge we propose contribute to the results; (iii) scales efficiently.
- Subjects :
- probability distribution
Computer science
ancestor source class
visual databases
target object appearance
Context (language use)
large-scale database
target object location
Machine learning
computer.software_genre
sibling source class
target object context
object class
large-scale knowledge transfer
Prototypes
Training
bounding-boxes
Visualization
Structure (mathematical logic)
Class (computer programming)
Support vector machines
business.industry
ImageNet
Context
object detection
statistical distributions
Vehicles
knowledge management
Airplanes
Object (computer science)
Object detection
Support vector machine
Artificial intelligence
business
computer
object localization
Subjects
Details
- Database :
- OpenAIRE
- Journal :
- 2012 IEEE Conference on Computer Vision and Pattern Recognition
- Accession number :
- edsair.doi.dedup.....2c74caefe98f85e7387d885943aaeed9