The novel coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused a major pandemic outbreak recently. Various diagnostic technologies have been under active development. The novel coronavirus disease (COVID-19) may induce pulmonary failures, and chest X-ray imaging becomes one of the major confirmed diagnostic technologies. The very limited number of publicly available samples has rendered the training of the deep neural networks unstable and inaccurate. This study proposed a two-step transfer learning pipeline and a deep residual network framework COVID19XrayNet for the COVID-19 detection problem based on chest X-ray images. COVID19XrayNet firstly tunes the transferred model on a large dataset of chest X-ray images, which is further tuned using a small dataset of annotated chest X-ray images. The final model achieved 0.9108 accuracy. The experimental data also suggested that the model may be improved with more training samples being released. COVID19XrayNet, a two-step transfer learning framework designed for biomedical images.

/pdf/covid19xraynet-a-two-step-transfer-learning-model-for-the-4t51ylekuj.pdf

COVID19XrayNet: A Two-Step Transfer Learning Model for the COVID-19 Detecting Problem Based on a Limited Number of Chest X-Ray Images.

Labeling objects at a subordinate level typically requires expert knowledge, which is not always available when using random annotators. As such, learning directly from web images for fine-grained recognition has attracted broad attention. However, the presence of label noise and hard examples in web images are two obstacles for training robust fine-grained recognition models. Therefore, in this paper, we propose a novel approach for removing irrelevant samples from real-world web images during training, while employing useful hard examples to update the network. Thus, our approach can alleviate the harmful effects of irrelevant noisy web images and hard examples to achieve better performance. Extensive experiments on three commonly used fine-grained datasets demonstrate that our approach is far superior to current state-of-the-art web-supervised methods. The data and source code of this work have been made publicly available at: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/NUST-Machine-Intelligence-Laboratory/Advanced-Softly-Update-Drop</uri> . 

Exploiting Web Images for Fine-Grained Visual Recognition by Eliminating Open-Set Noise and Utilizing Hard Examples

We proposed a Prior Gradient Mask Guided Pruning-aware Fine-Tuning (PGMPF) framework to accelerate deep Convolutional Neural Networks (CNNs). In detail, the proposed PGMPF selectively suppresses the gradient of those ”unimportant” parameters via a prior gradient mask generated by the pruning criterion during fine-tuning. PGMPF has three charming characteristics over previous works: (1) Pruning-aware network fine-tuning. A typical pruning pipeline consists of training, pruning and fine-tuning, which are relatively independent, while PGMPF utilizes a variant of the pruning mask as a prior gradient mask to guide fine-tuning, without
 complicated pruning criteria. (2) An excellent tradeoff between large model capacity during fine-tuning and stable convergence speed to obtain the final compact model. Previous works preserve more training information of pruned parameters during fine-tuning to pursue better performance, which would incur catastrophic non-convergence of the pruned model for relatively large pruning rates, while our PGMPF greatly stabilizes the fine-tuning phase by gradually constraining the learning rate of those ”unimportant” parameters. (3) Channel-wise random dropout of the prior gradient mask to impose some gradient noise to fine-tuning to further improve the robustness of final compact model. Experimental results on three image classification benchmarks CIFAR10/ 100 and ILSVRC-2012 demonstrate the effectiveness of our method for various CNN architectures, datasets and pruning rates. Notably, on ILSVRC-2012, PGMPF reduces 53.5% FLOPs on ResNet-50 with only 0.90% top-1 accuracy drop and 0.52% top-5 accuracy drop, which has advanced the state-of-the-art with negligible extra computational cost.

/pdf/prior-gradient-mask-guided-pruning-aware-fine-tuning-3b3wv6b7.pdf

Prior Gradient Mask Guided Pruning-Aware Fine-Tuning

We propose a fast partial video copy detection framework in this paper. In this framework all frame features of the reference videos are organized in a KNN searchable database. Instead of scanning all reference videos, the query video segment does a fast KNN search in the global feature database. The returned results are used to generate a short list of candidate videos. A modified temporal network is then used to localize the copy segment in the candidate videos. We evaluate different choice of CNN features on the VCDB dataset. Our benchmark F1 score exceeds the state of the art by a big margin.

A Fast Partial Video Copy Detection Using KNN and Global Feature Database.

Unlike in most previous partial video copy detection (PVCD) algorithms, where reference videos are scanned one by one, we treat the PVCD as a video search/retrieval problem. We propose a fast partial video copy detection framework in this paper. In this framework, all frame CNN features of the reference videos are organized in a KNN searchable database. Instead of scanning all reference videos, the query video segment does a fast KNN search in the global feature database. The returned results are used to generate a shortlist of candidate videos. A modified temporal network is then used to localize the copy segment in the candidate videos. Furthermore, We propose to use a transformer encoder to improve the CNN feature. We evaluate our algorithm on the VCDB dataset. Our benchmark F1 scores exceed state-of-the-art by a big margin. The speed of our algorithm is also improved significantly. 

/pdf/a-fast-partial-video-copy-detection-using-knn-and-global-2ez6r5yh.pdf

A Fast Partial Video Copy Detection Using KNN and Global Feature Database

Despite significant progress of applying deep learning methods to the field of content-based image retrieval, there has not been a software library that covers these methods in a unified manner. In order to fill this gap, we introduce PyRetri, an open source library for deep learning based unsupervised image retrieval. The library encapsulates the retrieval process in several stages and provides functionality that covers various prominent methods for each stage. The idea underlying its design is to provide a unified platform for deep learning based image retrieval research, with high usability and extensibility. The project source code, with usage examples, sample data and pre-trained models are available at https://github.com/PyRetri/.

PyRetri: A PyTorch-based Library for Unsupervised Image Retrieval by Deep Convolutional Neural Networks

Despite significant progress of applying deep learning methods to the field of content-based image retrieval, there has not been a software library that covers these methods in a unified manner. In order to fill this gap, we introduce PyRetri, an open source library for deep learning based unsupervised image retrieval. The library encapsulates the retrieval process in several stages and provides functionality that covers various prominent methods for each stage. The idea underlying its design is to provide a unified platform for deep learning based image retrieval research, with high usability and extensibility. To the best of our knowledge, this is the first open-source library for unsupervised image retrieval by deep learning.

Yuehu Liu

Papers

PyRetri: A PyTorch-based Library for Unsupervised Image Retrieval by Deep Convolutional Neural Networks

PyRetri: A PyTorch-based Library for Unsupervised Image Retrieval by Deep Convolutional Neural Networks