Learning with Side Information through Modality Hallucination

被引：146

作者：

Hoffman, Judy ^{[1
]}

Gupta, Saurabh ^{[1
]}

Darrell, Trevor ^{[1
]}

机构：

[1] Univ Calif Berkeley, EECS Dept, Berkeley, CA 94720 USA

来源：

2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2016年

关键词：

D O I：

10.1109/CVPR.2016.96

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present a modality hallucination architecture for training an RGB object detection model which incorporates depth side information at training time. Our convolutional hallucination network learns a new and complementary RGB image representation which is taught to mimic convolutional mid-level features from a depth network. At test time images are processed jointly through the RGB and hallucination networks to produce improved detection performance. Thus, our method transfers information commonly extracted from depth training data to a network which can extract that information from the RGB counterpart. We present results on the standard NYUDv2 dataset and report improvement on the RGB detection task.

引用

页码：826 / 834

页数：9