In this paper, we propose a new approach to learn multimodal multilingual embeddings for matching images and their relevant captions in two languages. We combine two existing objective functions to make images and captions close in a joint embedding space while adapting the alignment of word embeddings between existing languages in our model. We show that our approach enables better generalization, achieving state-of-the-art performance in text-to-image and image-to-text retrieval task, and caption-caption similarity task. Two multimodal multilingual datasets are used for evaluation: Multi30k with German and English captions and Microsoft-COCO with English and Japanese captions.
We propose a joint transmission-recognition scheme for efficient inference at the wireless network edge. Our scheme allows for reliable image recognition over wireless channels with significant computational load reduction at the sender side. We incorporate recently proposed deep joint source-channel coding (JSCC) scheme, and combine it with novel filter pruning strategies aimed at reducing the redundant complexity from neural networks. We evaluate our approach on a classification task, and show satisfactory results in both transmission reliability and workload reduction. This is the first work that combines deep JSCC with network pruning and applies it to images classification over wireless network.
We introduce an open-source toolkit, i.e., the deep Self End-to-end Learning Framework (deepSELF), as a toolkit of deep self end-to-end learning framework for multi-modal signals. To the best of our knowledge, it is the first public toolkit assembling a series of state-of-the-art deep learning technologies. Highlights of the proposed deepSELF toolkit include: First, it can be used to analyse a variety of multi-modal signals, including images, audio, and single or multi-channel sensor data. Second, we provide multiple options for pre-processing, e.g., filtering, or spectrum image generation by Fourier or wavelet transformation. Third, plenty of topologies in terms of NN, 1D/2D/3D CNN, and RNN/LSTM/GRU can be customised and a series of pretrained 2D CNN models, e.g., AlexNet, VGGNet, ResNet can be used easily. Last but not least, above these features, deepSELF can be flexibly used not only as a single model but also as a fusion of such.
As the performance and popularity of deep neural networks has increased, so too has their computational cost. There are many effective techniques for reducing a network's computational footprint (quantisation, pruning, knowledge distillation), but these lead to models whose computational cost is the same regardless of their input. Our human reaction times vary with the complexity of the tasks we perform: easier tasks (e.g. telling apart dogs from boat) are executed much faster than harder ones (e.g. telling apart two similar looking breeds of dogs). Driven by this observation, we develop a method for adaptive network complexity by attaching a small classification layer, which we call SideNet, to a large pretrained network, which we call MainNet. Given an input, the SideNet returns a classification if its confidence level, obtained via softmax, surpasses a user determined threshold, and only passes it along to the large MainNet for further processing if its confidence is too low. This allows us to flexibly trade off the network's performance with its computational cost. Experimental results show that simple single hidden layer perceptron SideNets added onto pretrained ResNet and BERT MainNets allow for substantial decreases in compute with minimal drops in performance on image and text classification tasks. We also highlight three other desirable properties of our method, namely that the classifications obtained by SideNets are calibrated, complementary to other compute reduction techniques, and that they enable the easy exploration of compute accuracy space.
Person re-identification aims at identifying a certain pedestrian across non-overlapping camera networks. Video-based re-identification approaches have gained significant attention recently, expanding image-based approaches by learning features from multiple frames. In this work, we propose a novel Co-Saliency Spatio-Temporal Interaction Network (CSTNet) for person re-identification in videos. It captures the common salient foreground regions among video frames and explores the spatial-temporal long-range context interdependency from such regions, towards learning discriminative pedestrian representation. Specifically, multiple co-saliency learning modules within CSTNet are designed to utilize the correlated information across video frames to extract the salient features from the task-relevant regions and suppress background interference. Moreover, multiple spatialtemporal interaction modules within CSTNet are proposed, which exploit the spatial and temporal long-range context interdependencies on such features and spatial-temporal information correlation, to enhance feature representation. Extensive experiments on two benchmarks have demonstrated the effectiveness of the proposed method.
In photography, low depth of field (DOF) is an important technique to emphasize the object of interest (OOI) within an image. Thus, low DOF images are widely used in the application area of macro, portrait or sports photography. When viewing a low DOF image, the viewer implicitly concentrates on the regions that are sharper regions of the image and thus segments the image into regions of interest and non regions of interest which has a major impact on the perception of the image. Thus, a robust algorithm for the fully automatic detection of the OOI in low DOF images provides valuable information for subsequent image processing and image retrieval. In this paper we propose a robust and parameterless algorithm for the fully automatic segmentation of low DOF images. We compare our method with three similar methods and show the superior robustness even though our algorithm does not require any parameters to be set by hand. The experiments are conducted on a real world data set with high and low DOF images.
We propose a 2D generalization to the $M$-band case of the dual-tree decomposition structure (initially proposed by N. Kingsbury and further investigated by I. Selesnick) based on a Hilbert pair of wavelets. We particularly address (\textit{i}) the construction of the dual basis and (\textit{ii}) the resulting directional analysis. We also revisit the necessary pre-processing stage in the $M$-band case. While several reconstructions are possible because of the redundancy of the representation, we propose a new optimal signal reconstruction technique, which minimizes potential estimation errors. The effectiveness of the proposed $M$-band decomposition is demonstrated via denoising comparisons on several image types (natural, texture, seismics), with various $M$-band wavelets and thresholding strategies. Significant improvements in terms of both overall noise reduction and direction preservation are observed.
Purpose: To develop and validate computer-aided diagnosis (CXDx) system for classification between COVID-19 pneumonia, non-COVID-19 pneumonia, and the healthy on chest X-ray (CXR) images. Because CXR datasets related with COVID-19 were small, transfer learning with pretrained models and combination of data augmentation methods were used to improve accuracy and robustness of the CADx system. Materials and Methods: From two public datasets, 1248 CXR images were obtained, which included 215, 533, and 500 CXR images of COVID-19 pneumonia, non-COVID-19 pneumonia, and the healthy. The proposed CADx system utilized VGG16 as a pretrained model and combination of conventional method and mixup as data augmentation methods. Other types of pretrained models were used for comparison with the VGG16-based model. In addition, single type or no data augmentation methods were also evaluated. Splitting of training/validation/test sets was used when building and evaluating the CADx system. Three-category accuracy was evaluated for test set with 125 CXR images. Results: The three-category accuracy of the CAD system was 83.6% between COVID-19 pneumonia, non-COVID-19 pneumonia, and the healthy. In addition, sensitivity of COVID-19 pneumonia was more than 90%. The combination of conventional method and mixup was more useful than single type or no data augmentation methods. Conclusions: It was possible to build the accurate CADx system for the 3-category classification of COVID-19 pneumonia, non-COVID-19 pneumonia, and the healthy.
Generative Adversarial Networks (GANs) have made great progress in synthesizing realistic images in recent years. However, they are often trained on image datasets with either too few samples or too many classes belonging to different data distributions. Consequently, GANs are prone to underfitting or overfitting, making the analysis of them difficult and constrained. Therefore, in order to conduct a thorough study on GANs while obviating unnecessary interferences introduced by the datasets, we train them on artificial datasets where there are infinitely many samples and the real data distributions are simple, high-dimensional and have structured manifolds. Moreover, the generators are designed such that optimal sets of parameters exist. Empirically, we find that under various distance measures, the generator fails to learn such parameters with the GAN training procedure. We also find that training mixtures of GANs leads to more performance gain compared to increasing the network depth or width when the model complexity is high enough. Our experimental results demonstrate that a mixture of generators can discover different modes or different classes automatically in an unsupervised setting, which we attribute to the distribution of the generation and discrimination tasks across multiple generators and discriminators. As an example of the generalizability of our conclusions to realistic datasets, we train a mixture of GANs on the CIFAR-10 dataset and our method significantly outperforms the state-of-the-art in terms of popular metrics, i.e., Inception Score (IS) and Fr\'echet Inception Distance (FID).
In remote sensing images, the presence of thick cloud accompanying cloud shadow is a high probability event, which can affect the quality of subsequent processing and limit the scenarios of application. Hence, removing the thick cloud and cloud shadow as well as recovering the cloud-contaminated pixels is indispensable to make good use of remote sensing images. In this paper, a novel thick cloud removal method for remote sensing images based on temporal smoothness and sparsity-regularized tensor optimization (TSSTO) is proposed. The basic idea of TSSTO is that the thick cloud and cloud shadow are not only sparse but also smooth along the horizontal and vertical direction in images while the clean images are smooth along the temporal direction between images. Therefore, the sparsity norm is used to boost the sparsity of the cloud and cloud shadow, and unidirectional total variation (UTV) regularizers are applied to ensure the unidirectional smoothness. This paper utilizes alternation direction method of multipliers to solve the presented model and generate the cloud and cloud shadow element as well as the clean element. The cloud and cloud shadow element is purified to get the cloud area and cloud shadow area. Then, the clean area of the original cloud-contaminated images is replaced to the corresponding area of the clean element. Finally, the reference image is selected to reconstruct details of the cloud area and cloud shadow area using the information cloning method. A series of experiments are conducted both on simulated and real cloud-contaminated images from different sensors and with different resolutions, and the results demonstrate the potential of the proposed TSSTO method for removing cloud and cloud shadow from both qualitative and quantitative viewpoints.