Authors:
(1) Hyeongjun Kwon, Yonsei University;
(2) Jinhyun Gang, University of Yunusi;
(3) Jin Kim, University of Uniony;
(4) Kwonyoung Kim, Yonsei University;
(5) Kwanghoon Sohn, Yonsei University and the Korea Institute of Science and Technology (KIST).
Links table
Abstract and 1 introduction
2. Related work
3. Excess engineering
4. The method
4.1. summary
4.2. Tree of possibly hierarchy
4.3. The optical hierarchy analyzes
4.4. Learn the hierarchy in the Zaidi space
4.5. Optical hierarchy coding
5. Experiences and 5.1. Classification of photos
5.2. Discovering organisms and retail the similar
5.3. Semantic
5.4. Perception
6. Huritage and discussion studies
7. Conclusion and references
A. Network structure
Theoretical foundation line
Additional results
D. Additional visualization
In this document, we include complementary materials “to improve visual identification with the appointment of the pyramid hierarchy.” We first provide more concrete implementation details (Article A), the theoretical foundation line (Article B), and additional experimental results (Article C). Finally, we imagine more visual hierarchical trees from the specific images to provide solid -style solid evidence (Article D).
A. Network structure
A.1. classification
A.2. Thick prediction.
Theoretical foundation line
B.1. A mixture of gaussians
![Table 6. Performance comparisons between full training and control across DNNS on the Imagenet-1K data set [36].](https://hackernoon.imgix.net/images/fWZa4tUiBGemnqQfBGgCPf9594N2-80931zo.png?auto=format&fit=max&w=1200)
This paper is available on arxiv under CC with a verb license 4.0.
