期刊
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
卷 25, 期 12, 页码 3244-3257出版社
IEEE COMPUTER SOC
DOI: 10.1109/TVCG.2018.2866793
关键词
Three-dimensional displays; Shape; Solid modeling; Estimation; Visualization; Task analysis; Computational modeling; 3D shape classification; multi-view 3D shape recognition; visual attention model; recurrent neural network; reinforcement learning; convolutional neural network
资金
- National Natural Science Foundation of China [61373135, 61672299, 61702281, 61532003, 61572507, 61622212]
- Postdoctoral Science Foundation of Jiangsu Province of China [1701046A]
Multi-view deep neural network is perhaps the most successful approach in 3D shape classification. However, the fusion of multi-view features based on max or average pooling lacks a view selection mechanism, limiting its application in, e.g., multi-view active object recognition by a robot. This paper presents VERAM, a view-enhanced recurrent attention model capable of actively selecting a sequence of views for highly accurate 3D shape classification. VERAM addresses an important issue commonly found in existing attention-based models, i.e., the unbalanced training of the subnetworks corresponding to next view estimation and shape classification. The classification subnetwork is easily overfitted while the view estimation one is usually poorly trained, leading to a suboptimal classification performance. This is surmounted by three essential view-enhancement strategies: 1) enhancing the information flow of gradient backpropagation for the view estimation subnetwork, 2) devising a highly informative reward function for the reinforcement training of view estimation and 3) formulating a novel loss function that explicitly circumvents view duplication. Taking grayscale image as input and AlexNet as CNN architecture, VERAM with 9 views achieves instance-level and class-level accuracy of 95.5 and 95.3 percent on ModelNet10, 93.7 and 92.1 percent on ModelNet40, both are the state-of-the-art performance under the same number of views.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据