The growing demand for high-quality video processing in real-time applications demands efficient denoising techniques that can operate swiftly while maintaining visual fidelity. Conventional approaches often struggle to balance these competing requirements, especially when dealing with high-resolution video streams or resource-constrained environments. This study aims to develop methods for accelerating video data denoising using deep convolutional neural networks while maintaining acceptable output quality. We selected the popular FastDVDNet denoising network, which operates on a sliding window principle, as our baseline for comparison and a starting point for our research. This paper proposes several modifications of FastDVDNet that significantly enhance computational efficiency. We introduce four key optimizations: caching intermediate denoising results, reducing intermediate channels in input block, simplifying convolutional blocks, and halving the number of channels. We evaluated these modifications on the Set8 dataset and compared the results with the original model at various noise levels. Finally, we introduce LiteDVDNet, a fine-tuned version of FastDVDNet model that achieves the optimal balance between processing speed, and denoising performance. We developed two model variants: LiteDVDNet-32, which is 3× faster than the original model with only 0.18 dB average PSNR reduction, and the more lightweight LiteDVDNet-16, which delivers a 5× speed improvement at the cost of 0.61 dB average PSNR reduction. 

Self-supervised learning has emerged as an effective paradigm for learning universal feature representations from vast amounts of unlabeled data. It’s remarkable success in recent years has been demonstrated in both natural language processing and computer vision domains. Serving as a cornerstone of the development of large-scale models, self-supervised learning has propelled the advancement of machine intelligence to new heights. In this paper, we draw inspiration from Siamese Networks and Masked Autoencoders to propose a denoising self-distilling Masked Autoencoder model for Self-supervised learning. The model is composed of a Masked Autoencoder and a teacher network, which work together to restore input image blocks corrupted by random Gaussian noise. Our objective function incorporates both pixel-level loss and high-level feature loss, allowing the model to extract complex semantic features. We evaluated our proposed method on three benchmark datasets, namely Cifar-10, Cifar-100, and STL-10, and compared it with classical self-supervised learning techniques. The experimental results demonstrate that our pre-trained model achieves a slightly superior fine-tuning performance on the STL-10 dataset, surpassing MAE by 0.1%. Overall, our method yields comparable experimental results when compared to other masked image modeling methods. The rationale behind our designed architecture is validated through ablation experiments. Our proposed method can serve as a complementary technique within the existing series of self-supervised learning approaches for masked image modeling, with the potential to be applied to larger datasets.

The coronavirus pandemic has been going on since the year 2019, and the trend is still not abating. Therefore, it is particularly important to classify medical CT scans to assist in medical diagnosis. At present, Supervised Deep Learning algorithms have made a great success in the classification task of medical CT scans, but medical image datasets often require professional image annotation, and many research datasets are not publicly available. To solve this problem, this paper is inspired by the self-supervised learning algorithm MAE and uses the MAE model pre-trained on ImageNet to perform transfer learning on CT Scans dataset. This method improves the generalization performance of the model and avoids the risk of overfitting on small datasets. Through extensive experiments on the COVID-CT dataset and the SARS-CoV-2 dataset, we compare the SSL-based method in this paper with other state-of-the-art supervised learning-based pretraining methods. Experimental results show that our method improves the generalization performance of the model more effectively and avoids the risk of overfitting on small datasets. The model achieved almost the same accuracy as supervised learning on both test datasets. Finally, ablation experiments aim to fully demonstrate the effectiveness of our method and how it works.