Real-time object detection method based on improved YOLOv4-tiny

9 Nov 2020 • Zicong Jiang • Liquan Zhao • Shuaiyang Li • Yanfei Jia

The "You only look once v4"(YOLOv4) is one type of object detection methods in deep learning. YOLOv4-tiny is proposed based on YOLOv4 to simple the network structure and reduce parameters, which makes it be suitable for developing on the mobile and embedded devices... (read more)

PDF Abstract

Results from the Paper

Add Remove

Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

METHOD	TYPE
k-Means Clustering	Clustering
Global Average Pooling	Pooling Operations
Logistic Regression	Generalized Linear Models
Xavier Initialization	Initialization
Bottleneck Residual Block	Skip Connection Blocks
ReLU	Activation Functions
1x1 Convolution	Convolutions
Max Pooling	Pooling Operations
Batch Normalization	Normalization
Bottom-up Path Augmentation	Feature Extractors
Residual Block	Skip Connection Blocks
FPN	Feature Extractors
YOLOv3	Object Detection Models
Softmax	Output Functions
Residual Connection	Skip Connections
Darknet-53	Convolutional Neural Networks
Label Smoothing	Regularization
ResNet-D	Convolutional Neural Networks
CutMix	Image Data Augmentation
Tanh Activation	Activation Functions
Convolution	Convolutions
Average Pooling	Pooling Operations
PAFPN	Feature Extractors
Cosine Annealing	Learning Rate Schedules
Softplus	Activation Functions
Spatial Pyramid Pooling	Pooling Operations
Sigmoid Activation	Activation Functions
DropBlock	Regularization
Spatial Attention Module	Image Model Blocks
CSPDarknet53	Convolutional Neural Networks
Mish	Activation Functions
YOLOv4	Object Detection Models

METHOD

TYPE

k-Means Clustering