Crowd Counting On Ucf Qnrf

Metrics

MAE

Results

Performance results of various models on this benchmark

Model Name	MAE	Paper Title	Repository
Cascaded-MTL	252	CNN-based Cascaded Multi-task Learning of High-level Prior and Density Estimation for Crowd Counting
Resnet101	190	Deep Residual Learning for Image Recognition
CAN	107	Context-Aware Crowd Counting
Switch-CNN	228	Switching Convolutional Neural Network for Crowd Counting
CLIP-EBC (ResNet50)	80.5	CLIP-EBC: CLIP Can Count Accurately through Enhanced Blockwise Classification
GauNet (ResNet-50)	81.6	Rethinking Spatial Invariance of Convolutional Networks for Object Counting
Idrees et al.	132	Composition Loss for Counting, Density Map Estimation and Localization in Dense Crowds	-
M-SFANet	85.6	Encoder-Decoder Based Convolutional Neural Networks with Multi-Scale-Aware Modules for Crowd Counting
APGCC	80.1	Improving Point-based Crowd Counting and Localization Based on Auxiliary Point Guidance
Encoder-Decoder	270	SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
CSRNet-EBC	79.3	CLIP-EBC: CLIP Can Count Accurately through Enhanced Blockwise Classification
SGANet + CL	87.6	Crowd Counting via Segmentation Guided Attention Networks and Curriculum Loss
PSL-Net	85.5	Crowd Counting and Individual Localization Using Pseudo Square Label	-
DMCount-EBC (32, dynamic)	76.06	CLIP-EBC: CLIP Can Count Accurately through Enhanced Blockwise Classification
Densenet201	163	Densely Connected Convolutional Networks
SGANet	89.1	Crowd Counting via Segmentation Guided Attention Networks and Curriculum Loss
Idrees et al.	315	Multi-source Multi-scale Counting in Extremely Dense Crowd Images	-
MCNN	277	Single-Image Crowd Counting via Multi-Column Convolutional Neural Network
DMCount-EBC	77.2	CLIP-EBC: CLIP Can Count Accurately through Enhanced Blockwise Classification
DMCount-EBC (16, dynamic)	75.90	CLIP-EBC: CLIP Can Count Accurately through Enhanced Blockwise Classification

0 of 22 row(s) selected.