Hybrid Transformer and CNN Attention Network for Stereo Image Super-resolution

Cheng, Ming; Ma, Haoyu; Ma, Qiufang; Sun, Xiaopeng; Li, Weiqi; Zhang, Zhenyu; Sheng, Xuhan; Zhao, Shijie; Li, Junlin; Zhang, Li

doi:10.1109/cvprw59228.2023.00171

Cited by 21 publications

(2 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The convolutional layers extract features from the input image, which are then down sampled by the max-pooling layers. The collar of YOLOv7 is path aggregation network (PANet), a pyramidal attention network [26]. The PANet is designed to improve YOLOv7 precision by identifying long distance dependencies between features.…”

Section: Yolov7mentioning

confidence: 99%

Swin transformer adaptation into YOLOv7 for road damage detection

Irsal,

Utaminingrum,

Ogata

2024

Bulletin EEI

View full text Add to dashboard Cite

Highways are an important component of any country. However, some highways in Indonesia endanger users while maintaining road safety. Crack detection early in the deterioration process can prevent further damage and lower maintenance costs. A recent study sought to develop a method for detecting road damage by combining the road damage detection (RDD) dataset with generative adversarial network technology and data augmentation to improve training. The current study aims to broaden the you only look once (YOLO) framework by incorporating the Swin Transformer into the chiral stationary phases (CSP) component of YOLOv7, with the goal of improving object detection accuracy in a variety of visual scenarios. The study compares the performance of various object detection models with varying parameters and configurations, such as YOLOv5l, YOLOv6l, YOLOv7-tiny, YOLOv7, and YOLOv7x. YOLOv5l has 46 million parameters and 108 billion floating point operations per second (FLOPS), whereas YOLOv6l has 59.5 million parameters and 150 billion FLOPS. With 31 million parameters and 140 billion FLOPS, the YOLOv7-swin model performs best with mean average precision (mAP), mAP_0.50 of 0.47. and mAP_0.5:0.95 of 0.232. The experimental results show that our YOLOv7-swin model outperforms both YOLOv7x and YOLOv7-tiny. The proposed model significantly improves object detection accuracy while keeping complexity and performance in balance.

show abstract

Section: Yolov7mentioning

confidence: 99%

Swin transformer adaptation into YOLOv7 for road damage detection

Irsal,

Utaminingrum,

Ogata

2024

Bulletin EEI

View full text Add to dashboard Cite

show abstract

“…Four classification frameworks have been explored to classify bird song [8]. Reference [9] proposes a lightweight image super-resolution reconstruction network based on Transformer CNN to address the issues of high computational complexity and significant memory consumption in existing super-resolution reconstruction networks, making it more suitable for application in embedded terminals such as mobile platforms. Reference [10] introduces the Transformer architecture, which is shown to better fuse global information in the image compared to the CNN structure.…”

Section: Introductionmentioning

confidence: 99%

Research on bird sound recognition based on deep learning

Sui,

Yang,

Wang

et al. 2023

International Conference on Automation Control, Algorithm, and Intelligent Bionics (ACAIB 2023)

View full text Add to dashboard Cite

To address the problems of low classification accuracy in bird sound recognition algorithms, a bird sound recognition method based on FBank feature extraction is proposed. FBank features are extracted from bird audio data, trained with Transformer model, and confusion matrix of the model is obtained by adjusting learning rate and optimizing loss function. Tests have shown that the proposed method has good accuracy in identifying 10 types of bird sounds.

show abstract

OmniSSR: Zero-Shot Omnidirectional Image Super-Resolution Using Stable Diffusion Model

Li,

Sheng,

et al. 2024

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Hybrid Transformer and CNN Attention Network for Stereo Image Super-resolution

Cited by 21 publications

References 31 publications

Swin transformer adaptation into YOLOv7 for road damage detection

Swin transformer adaptation into YOLOv7 for road damage detection

Research on bird sound recognition based on deep learning

OmniSSR: Zero-Shot Omnidirectional Image Super-Resolution Using Stable Diffusion Model

Contact Info

Product

Resources

About