Mutual information maximizing GAN inversion for real face with identity preservation

Recent generative adversarial networks (GANs) have yielded remarkable performance in face image synthesis. GAN inversion embeds an image into the latent space of a pretrained generator, enabling it to be used for real face manipulation. However, current inversion approaches for real faces suffer the dilemma of initialization collapse and identity loss. In this paper, we propose a hierarchical GAN inversion for real faces with identity preservation based on mutual information maximization. We first use a facial domain guaranteed initialization to avoid the initialization collapse. Furthermore, we prove that maximizing the mutual information between inverted faces and their identities is equivalent to minimizing the distance between identity features from inverted and original faces. Optimization for real face inversion with identity preservation is implemented on this mutual information-maximizing constraint. Extensive experimental results show that our approach outperforms state-of-the-art solutions for inverting and editing real faces, particularly in terms of face identity preservation.     

Collaborative Distribution Alignment for 2D image-based 3D shape retrieval

Retrieving 3D shapes with 2D images has become a popular research area nowadays, and a great deal of work has been devoted to reducing the discrepancy between 3D shapes and 2D images to improve retrieval performance. However, most approaches ignore the semantic information and decision boundaries of the two domains, and cannot achieve both domain alignment and category alignment in one module. In this paper, a novel Collaborative Distribution Alignment (CDA) model is developed to address the above existing challenges. Specifically, we first adopt a dual-stream CNN, following a similarity guided constraint module, to generate discriminative embeddings for input 2D images and 3D shapes (described as multiple views). Subsequently, we explicitly introduce a joint domain-class alignment module to dynamically learn a class-discriminative and domain-agnostic feature space, which can narrow the distance between 2D image and 3D shape instances of the same underlying category, while pushing apart the instances from different categories. Furthermore, we apply a decision boundary refinement module to avoid generating class-ambiguity embeddings by dynamically adjusting inconsistencies between two discriminators. Extensive experiments and evaluations on two challenging benchmarks, MI3DOR and MI3DOR-2, demonstrate the superiority of the proposed CDA method for 2D image-based 3D shape retrieval task.     

Multi-scale and multi-patch transformer for sandstorm image enhancement

Sandstorm is a meteorological phenomenon common in arid and semi-arid regions. A sandstorm can carry large volumes of sand unexpectedly, which leads to severe color deviations and significantly degraded visibility when an image is taken in such a scenario. However, existing image enhancement methods cannot enhance sandstorm images well due to the challenging degradations and the scarcity of sandstorm training data. In this paper, we propose a Transformer with rotary position embedding to perform sandstorm image enhancement via building multi-scale and multi-patch dependencies. Our key insights in this work are 1) a multi-scale Transformer can globally eliminate the color deviations of sandstorm images via aggregating global information, 2) a multi-patch Transformer can recover local details well via learning the spatial variant degradations, and 3) a U-shape Transformer with rotary position embedding as the core unit of multi-scale and multi-patch Transformer can effectively build the long-range dependencies. We also contribute a real-world Sandstorm Image Enhancement (SIE) dataset including 1,400 sandstorm images with different degrees of degradations and various scenes. Experiments performed on synthetic images and real-world sandstorm images demonstrate that our proposed method not only obtains visually pleasing results but also outperforms state-of-the-art methods qualitatively and quantitatively.     

Influencer-defaulter mutation-based optimization algorithms for predicting electricity prices

Efficient electricity price forecasting plays a significant role in our society. In this paper, a novel influencer-defaulter mutation (IDM) mutation operator has been proposed. The IDM operator has been combined with six well-known optimization algorithms to create mutated optimization algorithms whose performance has been tested on twenty-four standard benchmark functions. Further, the artificial neural network is integrated with mutated optimization algorithms to solve the electricity price prediction problem. The policymakers can identify appropriate variables based on the predicted prices to help future market planning. The statistical results prove the efficacy of the IDM operator on the recent optimization algorithms.     

Priority-MECE: A Mobile Edge Cloud Ecosystem Based on Priority Tasks Offloading

Mobile Networks and Applications - With the rapid development of Internet of things, the traditional city model is no longer applicable. Therefore, the emerging concept of smart city meets the...     

Software Optimization in Ultrasound Imaging Technique Using Improved Deep Belief Learning Network on the Internet of Medical Things Platform

Wireless Personal Communications - The Internet of Medical Things (IoMT) is the array of medical instruments and related technologies that link Information Technology (IT) systems in...     

Live video streaming service with pay-as-you-use model on Ethereum Blockchain and InterPlanetary file system

Wireless Networks - In centralized video streaming platforms, the platform owner, rather than the content producer, controls most of the content uploaded on the centralized video...     

Adaptive reversible data hiding for JPEG images with multiple two-dimensional histograms

Joint photographic experts group (JPEG) can provide good quality with small file size but also eliminate extensively the redundancies of images. Therefore, hiding data into JPEG images in terms of maintaining high visual quality at small file sizes has been a great challenge for researchers. In this paper, an adaptive reversible data hiding method for JPEG images containing multiple two-dimensional (2D) histograms is proposed. Adaptability is mainly reflected in three aspects. The first one is to preferentially select sharper histograms for data embedding after $K$ histograms are established by constructing the $k$th ($k\in \{1,2,\dots ,K\}$) histogram using the $k$th non-zero alternating current (AC) coefficient of all the quantized discrete cosine transform blocks. On the other hand, to fully exploit the strong correlation between coefficients of one histogram, the smoothness of each coefficient is estimated by a block smoothness estimator so that a sharply-distributed 2D-histogram is constructed by combining two coefficients with similar smoothness into a pair. The pair corresponding to low complexity is selected priorly for data embedding, leading to high embedding performance while maintaining low file size. Besides, we design multiple embedding strategies to adaptively select the embedding strategy for each 2D histogram. Experimental results demonstrate that the proposed method can achieve higher rate–distortion performance which maintaining lower file storage space, compared with previous studies.     

Decomposition and replacement: Spatial knowledge distillation for monocular depth estimation

Knowledge distillation has become a key technique for making smart and light-weight networks through model compression and transfer learning. Unlike previous methods that applied knowledge distillation to the classification task, we propose to exploit the decomposition-and-replacement based distillation scheme for depth estimation from a single RGB color image. To do this, Laplacian pyramid-based knowledge distillation is firstly presented in this paper. The key idea of the proposed method is to transfer the rich knowledge of the scene depth, which is well encoded through the teacher network, to the student network in a structured way by decomposing it into the global context and local details. This is fairly desirable for the student network to restore the depth layout more accurately with limited resources. Moreover, we also propose a new guidance concept for knowledge distillation, so-called ReplaceBlock, which replaces blocks randomly selected in the decoded feature of the student network with those of the teacher network. Our ReplaceBlock gives a smoothing effect in learning the feature distribution of the teacher network by considering the spatial contiguity in the feature space. This process is also helpful to clearly restore the depth layout without the significant computational cost. Based on various experimental results on benchmark datasets, the effectiveness of our distillation scheme for monocular depth estimation is demonstrated in details. The code and model are publicly available at : https://github.com/tjqansthd/Lap_Rep_KD_Depth.