Weakly Supervised Localization Using Deep Feature Maps
العنوان: | Weakly Supervised Localization Using Deep Feature Maps |
---|---|
المؤلفون: | S. Karthikeyan, Heesung Kwon, Archith J. Bency, B.S. Manjunath, Hyungtae Lee |
المصدر: | Computer Vision – ECCV 2016 ISBN: 9783319464473 ECCV (1) |
بيانات النشر: | Springer International Publishing, 2016. |
سنة النشر: | 2016 |
مصطلحات موضوعية: | Class (computer programming), business.industry, Cognitive neuroscience of visual object recognition, Pattern recognition, Scale (descriptive set theory), 02 engineering and technology, Object (computer science), Convolutional neural network, Image (mathematics), Feature (computer vision), 020204 information systems, 0202 electrical engineering, electronic engineering, information engineering, Beam search, 020201 artificial intelligence & image processing, Computer vision, Artificial intelligence, business, Mathematics |
الوصف: | Object localization is an important computer vision problem with a variety of applications. The lack of large scale object-level annotations and the relative abundance of image-level labels makes a compelling case for weak supervision in the object localization task. Deep Convolutional Neural Networks are a class of state-of-the-art methods for the related problem of object recognition. In this paper, we describe a novel object localization algorithm which uses classification networks trained on only image labels. This weakly supervised method leverages local spatial and semantic patterns captured in the convolutional layers of classification networks. We propose an efficient beam search based approach to detect and localize multiple objects in images. The proposed method significantly outperforms the state-of-the-art in standard object localization data-sets. |
ردمك: | 978-3-319-46447-3 |
URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_________::645778275b3fbf2c08474b1e20aadc5c https://doi.org/10.1007/978-3-319-46448-0_43 |
حقوق: | OPEN |
رقم الأكسشن: | edsair.doi...........645778275b3fbf2c08474b1e20aadc5c |
قاعدة البيانات: | OpenAIRE |
ردمك: | 9783319464473 |
---|