Local Relation Networks for Image Recognition 英文详解
2022/3/19 23:31:02
本文主要是介绍Local Relation Networks for Image Recognition 英文详解,对大家解决编程问题具有一定的参考价值,需要的程序猿们随着小编来一起学习吧!
Local Relation Network
Adapt filter according to the appearance affinity
Motivation
Meaningful and adaptive spatical aggregation
Humans have a remarkable ability to “see the infinite world with finite means” [26, 2].
- Recognition-by-components: a theory of human image understanding.
- W. von Humboldt. On Language: On the Diversity of Human Language Construction and Its Influence on the Mental Development of the Human Species. Cambridge Texts in the History of Philosophy. Cambridge University Press, 1999/1836.
Hierarchical features -> different levels of features
Rather than recognizing how elements can be meaningfully joined together, convolutional layers act as templates
1 filter -> 1 channel
it'a waste of channels.
local relation layer
locality
&geometric priors
- determine feature buttom up
Convolution Layers and its Evolution
-
accuracy-efficiency trade-off
- group convolution
- depth-wise convolution
-
enlarge receptive field
- dialated convolution
deformable convolution
- active convolution
-
relax the requirement for sharing weights(this is too rigid)
locally connected layers
(DeepFace)
-
Capsule Networks
self-enhancement
filter bubble
Given that we prefer to eschew negative experiences, it comes as no surprise that people avoid the immediate psychological discomfort from cognitive dissonance by simply not reading or listening to differing opinions.
- Self-Attention/ Graph Neural Networks
- for
long-range
context
- for
This work
- a new
feature extractor
- introducing the
compositionality
directly intorepresention
Some concepts
bottom-up
&top-down
aggregation
-
geometric prior
-
locality
Algorithm
Local-Relation Networks are LR-nets
Suppose
\(C = 24, m = 8, k = 7,C/m = 3\)
We observe no accuracy drop with up to 8 channels (default) sharing the same aggregation(for k)
\(H = 160,W = 160\)
In this architecture, receptive field is relevant to the concept of
Geometry Prior
Or rather, learned Geometry Prior is used withneighborhood
(similar to receptive field.)k is the
neighborhood size
Geometry Prior is analogous to conventionalconvolution filter
However, geometry prior is considered together with appearance composability, which brings about adaption from input
In other words, the geometry prior is conditioned on the input pixels' correlation.
-
Input Feature Map
24x160x160
- 1x1 conv
- Query:
3x160*160
(compress #channels from 24 to 3)- 160x160 points, every point has a
query
inc/m = 3
channels.
- 160x160 points, every point has a
- Key:
3x160*160
- with kernel size
k = 7
, there are manyregions
in key maps
- with kernel size
- Query:
- 1x1 conv
- Geometry Prior:
3x7x7
- for every
region/neighbor
- for every
- Geometry Prior:
- 1x1 conv
-
\(W_{neighbour} = \text{SoftMax}(\text{Geo.}+\text{App.})\)(Geometry and Appearance)
-
\(\text{pixel}_{x,y} = W_{neighbour}\text{Input}_{neighbour}\)
- \(neighbor\) is kernel of size
k
centered atx,y
(the source and target pixel position.)
- \(neighbor\) is kernel of size
-
All of the aggregation is performed in a receptive field of
kxk
Design and Analysis
\[W = \text{SoftMax}(\text{GeoPrior}+\text{AppearanceComposability}) \]- Locality
They claim that LR
(i.e. Local-Relation Layer) can utilize large kernels more effectively
This difference may be due to the representation power of convolution layer being bottlenecked by the number of fixed filters, hence there is no benefit from a larger kernel size.
Weight Sharing across different positions in an image limits the utilization of the representation power of large kernels.
- Appearance composability
While in previous works the query and key are vectors, in the local relation layer, we use scalars to represent them so that the computation and representation are lightweight.
- Geometric Prior
What is that?
这篇关于Local Relation Networks for Image Recognition 英文详解的文章就介绍到这儿,希望我们推荐的文章对大家有所帮助,也希望大家多多支持为之网!
- 2024-05-29Elasticsearch慢查询日志配置
- 2024-05-29揭秘华为如此多成功项目的产品关键——Charter模板
- 2024-05-29海外IDC业务拓展的7大挑战
- 2024-05-29InLine Chat功能优化对标Github Copilot,CodeGeeX带来更高效、更直观的编程体验!
- 2024-05-29CodeGeeX 智能编程助手 6 项功能升级,在Visual Studio插件市场霸榜2周!
- 2024-05-29AutoMQ 生态集成 Apache Doris
- 2024-05-292024年IDC行业的深度挖掘:机遇、挑战与未来展望
- 2024-05-29五款扩展组件齐发 —— Volcano、Keda、Crane-scheduler 等,邀你体验
- 2024-05-29AutoMQ 对象存储数据高效组织的秘密: Compaction
- 2024-05-29活动预告|来 GIAC 大会听大数据降本利器:AutoMQ 基于云原生重新设计的 Kafka