DCTCNet:用于视觉识别的频率离散等号变换卷积网络
Jiayong Bao1, Jiangshe Zhang1, Chunxia Zhang1
1School of Mathematics and Statistics, Xi'an Jiaotong University, Xi'an 710049, China.
概括
本研究介绍了全相序数DCT卷积 (APSeDCTConv) 来增强计算机视觉模型. APSeDCTConv降低了计算成本,并提高了图像分类,对象检测和实例分割任务的性能.
科学领域:
- 计算机视觉 计算机视觉
- 深度学习 (Deep Learning) 是一种深度学习.
- 信号处理 信号处理
背景情况:
- 离散坐标变换 (DCT) 对于图像压缩和质量至关重要,但受到阻塞效应的影响.
- 传统DCT中的阻塞效应限制了其在计算机视觉应用中的效率.
- 解决这些局限性对于推进视觉任务至关重要.
研究的目的:
- 在卷积网络中引入一种新的DCT变体,即全相序数DCT (APSeDCT).
- 开发基于APSeDCT的卷积模块 (APSeDCTConv) 来提取多频信息.
- 提出使用APSeDCTConv的增强卷积运算符 (MultiConv),以提高模型效率和性能.
主要方法:
- APSeDCT被集成到卷积网络中,创建了APSeDCTConv,它模仿了香草卷积.
- 一个MultiConv操作员是通过增强APSeDCTConv设计的.
- 为了实验,ResNet的最后三个瓶块被MultiConv取代.
主要成果:
- APSeDCTConv显示了类似于标准卷曲的可转移性.
- 用MultiConv取代ResNet块可以降低计算成本和参数数量.
- 在ImageNet上的图像分类中,在各种模型 (ResNet,Res2Net,ResNext) 中观察到一致的性能增长.
- 对象检测和COCO上的实例细分分别显示了0.5%-1.1%和0.4%-0.7%的AP改进.
结论:
- APSeDCTConv有效地从深层特征中提取多频信息.
- 拟议的MultiConv操作器提高了计算机视觉任务中的模型效率和性能.
- APSeDCTConv增强为改善视觉应用中的深度学习模型提供了一个有希望的方法.
相关概念视频
01:24Convolution: Math, Graphics, and Discrete Signals
223
In any LTI (Linear Time-Invariant) system, the convolution of two signals is denoted using a convolution operator, assuming all initial conditions are zero. The convolution integral can be divided into two parts: the zero-input or natural response and the zero-state or forced response, with t0 indicating the initial time.
To simplify the convolution integral, it is assumed that both the input signal and impulse response are zero for negative time values. The graphical convolution process...
To simplify the convolution integral, it is assumed that both the input signal and impulse response are zero for negative time values. The graphical convolution process...
223
01:16Basic Discrete Time Signals
190
The unit step sequence is defined as 1 for zero and positive values of the integer n. This sequence can be graphically displayed using a set of eight sample points, showing a step function starting from n=0 and remaining constant thereafter.
The unit impulse or sample sequence is mathematically expressed as zero for all n values except at n=0, where it is one. The unit impulse sequence, denoted by δ(n), is the first difference of the unit step sequence, while the unit step sequence u(n) is...
The unit impulse or sample sequence is mathematically expressed as zero for all n values except at n=0, where it is one. The unit impulse sequence, denoted by δ(n), is the first difference of the unit step sequence, while the unit step sequence u(n) is...
190
01:17Convolution Properties II
168
The important convolution properties include width, area, differentiation, and integration properties.
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
168
01:20Convolution Properties I
136
Convolution computations can be simplified by utilizing their inherent properties.
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
The commutative property reveals that the input and the impulse response of an LTI (Linear Time-Invariant) system can be interchanged without affecting the output:
136
01:15Discrete Fourier Transform
215
The Discrete Fourier Transform (DFT) is a fundamental tool in signal processing, extending the discrete-time Fourier transform by evaluating discrete signals at uniformly spaced frequency intervals. This transformation converts a finite sequence of time-domain samples into frequency components, each representing complex sinusoids ordered by frequency. The DFT translates these sequences into the frequency domain, effectively indicating the magnitude and phase of each frequency component present...
215
01:26Visual System
498
Light enters the eye through the cornea, a transparent, dome-shaped surface covering the surface of the eyeball that helps to direct and focus incoming light. This light is then channeled toward the pupil, an adjustable opening whose size is controlled by the iris. The iris, a pigmented muscle, regulates the amount of light entering the eye by contracting or dilating the pupil, thereby ensuring optimal light levels for clear vision.
Once through the pupil, the light passes through the lens, a...
Once through the pupil, the light passes through the lens, a...
498