Transformer Fusion

Transformer fusion combines the strengths of convolutional neural networks (CNNs) and transformer networks to improve performance in various computer vision tasks, particularly image segmentation and object detection. Current research focuses on developing hybrid architectures that effectively integrate the local feature extraction capabilities of CNNs with the global context modeling of transformers, often employing strategies like early or late fusion and incorporating specialized attention mechanisms. These advancements yield significant improvements in accuracy and robustness, especially for challenging scenarios such as low-light conditions or the detection of small objects, impacting fields like autonomous driving, medical image analysis, and remote sensing.

Papers

July 3, 2024

Relating CNN-Transformer Fusion Network for Change Detection
Yuhao Gao, Gensheng Pei, Mengmeng Sheng, Zeren Sun, Tao Chen, Yazhou Yao
Convolutional Neural Network Change Detection Multi Scale Feature Fusion Transformer Fusion

April 27, 2024

CLFT: Camera-LiDAR Fusion Transformer for Semantic Segmentation in Autonomous Driving
Junyi Gu, Mauro Bellone, Tomáš Pivoňka, Raivo Sell
Vision Transformer Semantic Segmentation Autonomous Driving Vision Transformer Network Camera LiDAR Fusion LiDAR Fusion Transformer Fusion

January 10, 2024

Transformer-CNN Fused Architecture for Enhanced Skin Lesion Segmentation
Siddharth Tiwari
Convolutional Neural Network Medical Image Segmentation Skin Lesion Segmentation Convolution Operation Transformer Fusion

July 27, 2023

pCTFusion: Point Convolution-Transformer Fusion with Semantic Aware Loss for Outdoor LiDAR Point Cloud Segmentation
Abhishek Kuriyal, Vaibhav Kumar, Bharat Lohani
Point Cloud LiDAR Point Cloud Exceptional Point Point Cloud Segmentation Semantic Loss GTR CTRL Transformer Fusion

June 26, 2023

CST-YOLO: A Novel Method for Blood Cell Detection Based on Improved YOLOv7 and CNN-Swin Transformer
Ming Kang, Chee-Ming Ting, Fung Fung Ting, Raphaël Phan
Convolutional Layer Swin Transformer YOLOv5 Model New Method Blood Cell Transformer Fusion

January 17, 2023

Cooperation Learning Enhanced Colonic Polyp Segmentation Based on Transformer-CNN Fusion
Yuanyuan Wang, Zhaohong Deng, Qiongdan Lou, Shudong Hu, Kup-sze Choi, Shitong Wang
Deep Learning Feature Fusion Polyp Segmentation Fusion Transformer Transformer Fusion

Transformer Fusion

Papers

Relating CNN-Transformer Fusion Network for Change Detection

CLFT: Camera-LiDAR Fusion Transformer for Semantic Segmentation in Autonomous Driving

Transformer-CNN Fused Architecture for Enhanced Skin Lesion Segmentation

pCTFusion: Point Convolution-Transformer Fusion with Semantic Aware Loss for Outdoor LiDAR Point Cloud Segmentation

CST-YOLO: A Novel Method for Blood Cell Detection Based on Improved YOLOv7 and CNN-Swin Transformer

Cooperation Learning Enhanced Colonic Polyp Segmentation Based on Transformer-CNN Fusion