Главная
Study mode:
on
1
Introduction
2
Capsule Convention
3
Differences of SCN and CapsNet
4
Subspace Capsule Intuition
5
Subspace Capsule Network Principles
6
Subspace Capsules Based on Orthogonal Projection
7
Subspace Capsules in Intermediate Layers
8
Challenge
9
Proposed Method
10
No Information Loss
11
Norm preserving
12
Angle Preserving
13
Subspace Capsule Convolution using P
14
Activation Functions
15
Experiments
16
Datasets
17
Semi-supervised Image Classification
18
Semi-supervised classification with SCN-GAN
19
Image Generation with SCN-GAN
20
High Resolution Image Generation
21
Qualitative Results
22
Quantitative Comparison
23
Interpolation in the Latent Space
24
ImageNet Supervised Classification
25
Last Resnet Block Architecture
26
Effect of Capsule Size
27
Summery
Description:
Explore a 22-minute conference talk from AAAI 2020 presented by researchers from the University of Central Florida on Subspace Capsule Networks (SCN). Delve into the key differences between SCN and CapsNet, understand the intuition behind subspace capsules, and learn about their principles and implementation using orthogonal projection. Discover how subspace capsules function in intermediate layers and the challenges they address. Examine the proposed method, which ensures no information loss, norm preservation, and angle preservation. Investigate subspace capsule convolution using P-activation functions and review experimental results on various datasets, including semi-supervised image classification, image generation with SCN-GAN, and high-resolution image generation. Analyze qualitative and quantitative comparisons, explore interpolation in the latent space, and understand the impact of capsule size on ImageNet supervised classification performance.

Subspace Capsule Network

University of Central Florida
Add to list
0:00 / 0:00