CloudMix: Dual Mixup Consistency for Unpaired Point Cloud Completion.

Due to the unsatisfactory performance of supervised methods on unpaired real-world scans, point cloud completion via cross-domain adaptation has recently drawn growing attention. Nevertheless, previous approaches only focus on alleviating the distribution shift through domain alignment, resulting in massive information loss of real-world domain data. To tackle this issue, we propose a dual mixup-induced consistency regularization to integrate both source and target domain to improve robustness and generalization capability. Specifically, we mix up virtual and real-world shapes in the input and latent feature space respectively, and then regularize the completion network by forcing two kinds of mixed completion predictions to be consistent. To further adapt to each instance within the real-world domain, we design a novel density-aware refiner to utilize local context information to preserve the fine-grained details and remove noise or outliers for coarse completion. Extensive experiments on real-world scans and our synthetic unpaired datasets demonstrate the superiority of our method over existing state-of-the-art approaches.

GRA: Graph Representation Alignment for Semi-Supervised Action Recognition.

Graph convolutional networks (GCNs) have emerged as a powerful tool for action recognition, leveraging skeletal graphs to encapsulate human motion. Despite their efficacy, a significant challenge remains the dependency on huge labeled datasets. Acquiring such datasets is often prohibitive, and the frequent occurrence of incomplete skeleton data, typified by absent joints and frames, complicates the testing phase. To tackle these issues, we present graph representation alignment (GRA), a novel approach with two main contributions: 1) a self-training (ST) paradigm that substantially reduces the need for labeled data by generating high-quality pseudo-labels, ensuring model stability even with minimal labeled inputs and 2) a representation alignment (RA) technique that utilizes consistency regularization to effectively reduce the impact of missing data components. Our extensive evaluations on the NTU RGB+D and Northwestern-UCLA (N-UCLA) benchmarks demonstrate that GRA not only improves GCN performance in data-constrained environments but also retains impressive performance in the face of data incompleteness.

Self-Adversarial Disentangling for Specific Domain Adaptation.

Domain adaptation aims to bridge the domain shifts between the source and the target domain. These shifts may span different dimensions such as fog, rainfall, etc. However, recent methods typically do not consider explicit prior knowledge about the domain shifts on a specific dimension, thus leading to less desired adaptation performance. In this article, we study a practical setting called Specific Domain Adaptation (SDA) that aligns the source and target domains in a demanded-specific dimension. Within this setting, we observe the intra-domain gap induced by different domainness (i.e., numerical magnitudes of domain shifts in this dimension) is crucial when adapting to a specific domain. To address the problem, we propose a novel Self-Adversarial Disentangling (SAD) framework. In particular, given a specific dimension, we first enrich the source domain by introducing a domainness creator with providing additional supervisory signals. Guided by the created domainness, we design a self-adversarial regularizer and two loss functions to jointly disentangle the latent representations into domainness-specific and domainness-invariant features, thus mitigating the intra-domain gap. Our method can be easily taken as a plug-and-play framework and does not introduce any extra costs in the inference time. We achieve consistent improvements over state-of-the-art methods in both object detection and semantic segmentation.

Contrastive Learning for Joint Normal Estimation and Point Cloud Filtering.

Point cloud filtering and normal estimation are two fundamental research problems in the 3D field. Existing methods usually perform normal estimation and filtering separately and often show sensitivity to noise and/or inability to preserve sharp geometric features such as corners and edges. In this paper, we propose a novel deep learning method to jointly estimate normals and filter point clouds. We first introduce a 3D patch based contrastive learning framework, with noise corruption as an augmentation, to train a feature encoder capable of generating faithful representations of point cloud patches while remaining robust to noise. These representations are consumed by a simple regression network and supervised by a novel joint loss, simultaneously estimating point normals and displacements that are used to filter the patch centers. Experimental results show that our method well supports the two tasks simultaneously and preserves sharp features and fine details. It generally outperforms state-of-the-art techniques on both tasks.

3D Intracranial Aneurysm Classification and Segmentation via Unsupervised Dual-Branch Learning.

Intracranial aneurysms are common nowadays and how to detect them intelligently is of great significance in digital health. Whereas most existing deep learning research focused on medical images in a supervised way, we introduce an unsupervised method for the detection of intracranial aneurysms based on 3D point cloud data. In particular, our method consists of two stages: unsupervised pre-training and downstream tasks. As for the former, the main idea is to pair each point cloud with its jittering counterpart and maximise their correspondence. Then we design a dual-branch contrastive network with an encoder for each branch and a subsequent common projection head. As for the latter, we design simple networks for supervised classification and segmentation training. Experiments on the public dataset (IntrA) show that our unsupervised method achieves comparable or even better performance than some state-of-the-art supervised techniques, and it is most prominent in the detection of aneurysmal vessels. Experiments on the ModelNet-40 also show that our method achieves the accuracy of 90.79% which outperforms existing state-of-the-art unsupervised models.

AI-based carcinoma detection and classification using histopathological images: A systematic review.

Histopathological image analysis is the gold standard to diagnose cancer. Carcinoma is a subtype of cancer that constitutes more than 80% of all cancer cases. Squamous cell carcinoma and adenocarcinoma are two major subtypes of carcinoma, diagnosed by microscopic study of biopsy slides. However, manual microscopic evaluation is a subjective and time-consuming process. Many researchers have reported methods to automate carcinoma detection and classification. The increasing use of artificial intelligence (AI) in the automation of carcinoma diagnosis also reveals a significant rise in the use of deep network models. In this systematic literature review, we present a comprehensive review of the state-of-the-art approaches reported in carcinoma diagnosis using histopathological images. Studies are selected from well-known databases with strict inclusion/exclusion criteria. We have categorized the articles and recapitulated their methods based on specific organs of carcinoma origin. Further, we have summarized pertinent literature on AI methods, highlighted critical challenges and limitations, and provided insights on future research direction in automated carcinoma diagnosis. Out of 101 articles selected, most of the studies experimented on private datasets with varied image sizes, obtaining accuracy between 63% and 100%. Overall, this review highlights the need for a generalized AI-based carcinoma diagnostic system. Additionally, it is desirable to have accountable approaches to extract microscopic features from images of multiple magnifications that should mimic pathologists' evaluations.

Low Rank Matrix Approximation for 3D Geometry Filtering.

We propose a robust normal estimation method for both point clouds and meshes using a low rank matrix approximation algorithm. First, we compute a local isotropic structure for each point and find its similar, non-local structures that we organize into a matrix. We then show that a low rank matrix approximation algorithm can robustly estimate normals for both point clouds and meshes. Furthermore, we provide a new filtering method for point cloud data to smooth the position data to fit the estimated normals. We show the applications of our method to point cloud filtering, point set upsampling, surface reconstruction, mesh denoising, and geometric texture removal. Our experiments show that our method generally achieves better results than existing methods.

Explicit Facial Expression Transfer via Fine-Grained Representations.

Facial expression transfer between two unpaired images is a challenging problem, as fine-grained expression is typically tangled with other facial attributes. Most existing methods treat expression transfer as an application of expression manipulation, and use predicted global expression, landmarks or action units (AUs) as a guidance. However, the prediction may be inaccurate, which limits the performance of transferring fine-grained expression. Instead of using an intermediate estimated guidance, we propose to explicitly transfer facial expression by directly mapping two unpaired input images to two synthesized images with swapped expressions. Specifically, considering AUs semantically describe fine-grained expression details, we propose a novel multi-class adversarial training method to disentangle input images into two types of fine-grained representations: AU-related feature and AU-free feature. Then, we can synthesize new images with preserved identities and swapped expressions by combining AU-free features with swapped AU-related features. Moreover, to obtain reliable expression transfer results of the unpaired input, we introduce a swap consistency loss to make the synthesized images and self-reconstructed images indistinguishable. Extensive experiments show that our approach outperforms the state-of-the-art expression manipulation methods for transferring fine-grained expressions while preserving other attributes including identity and pose.

Blur Removal Via Blurred-Noisy Image Pair.

Complex blur such as the mixup of space-variant and space-invariant blur, which is hard to model mathematically, widely exists in real images. In this article, we propose a novel image deblurring method that does not need to estimate blur kernels. We utilize a pair of images that can be easily acquired in low-light situations: (1) a blurred image taken with low shutter speed and low ISO noise; and (2) a noisy image captured with high shutter speed and high ISO noise. Slicing the blurred image into patches, we extend the Gaussian mixture model (GMM) to model the underlying intensity distribution of each patch using the corresponding patches in the noisy image. We compute patch correspondences by analyzing the optical flow between the two images. The Expectation Maximization (EM) algorithm is utilized to estimate the parameters of GMM. To preserve sharp features, we add an additional bilateral term to the objective function in the M-step. We eventually add a detail layer to the deblurred image for refinement. Extensive experiments on both synthetic and real-world data demonstrate that our method outperforms state-of-the-art techniques, in terms of robustness, visual quality, and quantitative metrics.

Pointfilter: Point Cloud Filtering via Encoder-Decoder Modeling.

Point cloud filtering is a fundamental problem in geometry modeling and processing. Despite of significant advancement in recent years, the existing methods still suffer from two issues: 1) they are either designed without preserving sharp features or less robust in feature preservation; and 2) they usually have many parameters and require tedious parameter tuning. In this article, we propose a novel deep learning approach that automatically and robustly filters point clouds by removing noise and preserving their sharp features. Our point-wise learning architecture consists of an encoder and a decoder. The encoder directly takes points (a point and its neighbors) as input, and learns a latent representation vector which goes through the decoder to relate the ground-truth position with a displacement vector. The trained neural network can automatically generate a set of clean points from a noisy input. Extensive experiments show that our approach outperforms the state-of-the-art deep learning techniques in terms of both visual quality and quantitative error metrics. The source code and dataset can be found at https://github.com/dongbo-BUAA-VR/Pointfilter.

A Search History-Driven Offspring Generation Method for the Real-Coded Genetic Algorithm.

In evolutionary algorithms, genetic operators iteratively generate new offspring which constitute a potentially valuable set of search history. To boost the performance of offspring generation in the real-coded genetic algorithm (RCGA), in this paper, we propose to exploit the search history cached so far in an online style during the iteration. Specifically, survivor individuals over the past few generations are collected and stored in the archive to form the search history. We introduce a simple yet effective crossover model driven by the search history (abbreviated as SHX). In particular, the search history is clustered, and each cluster is assigned a score for SHX. In essence, the proposed SHX is a data-driven method which exploits the search history to perform offspring selection after the offspring generation. Since no additional fitness evaluations are needed, SHX is favorable for the tasks with limited budget or expensive fitness evaluations. We experimentally verify the effectiveness of SHX over 15 benchmark functions. Quantitative results show that our SHX can significantly enhance the performance of RCGA, in terms of both accuracy and convergence speed. Also, the induced additional runtime is negligible compared to the total processing time.

Multi-label zero-shot learning with graph convolutional networks.

The goal of zero-shot learning (ZSL) is to build a classifier that recognizes novel categories with no corresponding annotated training data. The typical routine is to transfer knowledge from seen classes to unseen ones by learning a visual-semantic embedding. Existing multi-label zero-shot learning approaches either ignore correlations among labels, suffer from large label combinations, or learn the embedding using only local or global visual features. In this paper, we propose a Graph Convolution Networks based Multi-label Zero-Shot Learning model, abbreviated as MZSL-GCN. Our model first constructs a label relation graph using label co-occurrences and compensates the absence of unseen labels in the training phase by semantic similarity. It then takes the graph and the word embedding of each seen (unseen) label as inputs to the GCN to learn the label semantic embedding, and to obtain a set of inter-dependent object classifiers. MZSL-GCN simultaneously trains another attention network to learn compatible local and global visual features of objects with respect to the classifiers, and thus makes the whole network end-to-end trainable. In addition, the use of unlabeled training data can reduce the bias toward seen labels and boost the generalization ability. Experimental results on benchmark datasets show that our MZSL-GCN competes with state-of-the-art approaches.

Structural basis for adenylation and thioester bond formation in the ubiquitin E1.

The ubiquitin (Ub) and Ub-like (Ubl) protein-conjugation cascade is initiated by E1 enzymes that catalyze Ub/Ubl activation through C-terminal adenylation, thioester bond formation with an E1 catalytic cysteine, and thioester bond transfer to Ub/Ubl E2 conjugating enzymes. Each of these reactions is accompanied by conformational changes of the E1 domain that contains the catalytic cysteine (Cys domain). Open conformations of the Cys domain are associated with adenylation and thioester transfer to E2s, while a closed conformation is associated with pyrophosphate release and thioester bond formation. Several structures are available for Ub E1s, but none has been reported in the open state before pyrophosphate release or in the closed state. Here, we describe the structures of Schizosaccharomyces pombe Ub E1 in these two states, captured using semisynthetic Ub probes. In the first, with a Ub-adenylate mimetic (Ub-AMSN) bound, the E1 is in an open conformation before release of pyrophosphate. In the second, with a Ub-vinylsulfonamide (Ub-AVSN) bound covalently to the catalytic cysteine, the E1 is in a closed conformation required for thioester bond formation. These structures provide further insight into Ub E1 adenylation and thioester bond formation. Conformational changes that accompany Cys-domain rotation are conserved for SUMO and Ub E1s, but changes in Ub E1 involve additional surfaces as mutational and biochemical analysis of residues within these surfaces alter Ub E1 activities.

Kinetic Analyses of the Siderophore Biosynthesis Inhibitor Salicyl-AMS and Analogues as MbtA Inhibitors and Antimycobacterial Agents.

There is a paramount need for expanding the drug armamentarium to counter the growing problem of drug-resistant tuberculosis. Salicyl-AMS, an inhibitor of salicylic acid adenylation enzymes, is a first-in-class antibacterial lead compound for the development of tuberculosis drugs targeting the biosynthesis of salicylic-acid-derived siderophores. In this study, we determined the Ki of salicyl-AMS for inhibition of the salicylic acid adenylation enzyme MbtA from Mycobacterium tuberculosis (MbtAtb), designed and synthesized two new salicyl-AMS analogues to probe structure-activity relationships (SAR), and characterized these two analogues alongside salicyl-AMS and six previously reported analogues in biochemical and cell-based studies. The biochemical studies included determination of kinetic parameters ( Kiapp, konapp, koff, and tR) and analysis of the mechanism of inhibition. For these studies, we optimized production and purification of recombinant MbtAtb, for which Km and kcat values were determined, and used the enzyme in conjunction with an MbtAtb-optimized, continuous, spectrophotometric assay for MbtA activity and inhibition. The cell-based studies provided an assessment of the antimycobacterial activity and postantibiotic effect of the nine MbtAtb inhibitors. The antimycobacterial properties were evaluated using a strain of nonpathogenic, fast-growing Mycobacterium smegmatis that was genetically engineered for MbtAtb-dependent susceptibility to MbtA inhibitors. This convenient model system greatly facilitated the cell-based studies by bypassing the methodological complexities associated with the use of pathogenic, slow-growing M. tuberculosis. Collectively, these studies provide new information on the mechanism of inhibition of MbtAtb by salicyl-AMS and eight analogues, afford new SAR insights for these inhibitors, and highlight several suitable candidates for future preclinical evaluation.

GPF: GMM-Inspired Feature-Preserving Point Set Filtering.

Point set filtering, which aims at reconstructing noise-free point sets from their corresponding noisy inputs, is a fundamental problem in 3D geometry processing. The main challenge of point set filtering is to preserve geometric features of the underlying geometry while at the same time removing the noise. State-of-the-art point set filtering methods still struggle with this issue: some are not designed to recover sharp features, and others cannot well preserve geometric features, especially fine-scale features. In this paper, we propose a novel approach for robust feature-preserving point set filtering, inspired by the Gaussian Mixture Model (GMM). Taking a noisy point set and its filtered normals as input, our method can robustly reconstruct a high-quality point set which is both noise-free and feature-preserving. Various experiments show that our approach can soundly outperform the selected state-of-the-art methods, in terms of both filtering quality and reconstruction accuracy.

A Robust Scheme for Feature-Preserving Mesh Denoising.

In recent years researchers have made noticeable progresses in mesh denoising, that is, recovering high-quality 3D models from meshes corrupted with noise (raw or synthetic). Nevertheless, these state of the art approaches still fall short for robustly handling various noisy 3D models. The main technical challenge of robust mesh denoising is to remove noise while maximally preserving geometric features. In particular, this issue becomes more difficult for models with considerable amount of noise. In this paper we present a novel scheme for robust feature-preserving mesh denoising. Given a noisy mesh input, our method first estimates an initial mesh, then performs feature detection, identification and connection, and finally, iteratively updates vertex positions based on the constructed feature edges. Through many experiments, we show that our approach can robustly and effectively denoise various input mesh models with synthetic noise or raw scanned noise. The qualitative and quantitative comparisons between our method and the selected state of the art methods also show that our approach can noticeably outperform them in terms of both quality and robustness.

Stable analogues of OSB-AMP: potent inhibitors of MenE, the o-succinylbenzoate-CoA synthetase from bacterial menaquinone biosynthesis.

MenE, the o-succinylbenzoate (OSB)-CoA synthetase from bacterial menaquinone biosynthesis, is a promising new antibacterial target. Sulfonyladenosine analogues of the cognate reaction intermediate, OSB-AMP, have been developed as inhibitors of the MenE enzymes from Mycobacterium tuberculosis (mtMenE), Staphylococcus aureus (saMenE) and Escherichia coli (ecMenE). Both a free carboxylate and a ketone moiety on the OSB side chain are required for potent inhibitory activity. OSB-AMS (4) is a competitive inhibitor of mtMenE with respect to ATP (K(i) =5.4±0.1 nM) and a noncompetitive inhibitor with respect to OSB (K(i) =11.2±0.9 nM). These data are consistent with a Bi Uni Uni Bi Ping-Pong kinetic mechanism for these enzymes. In addition, OSB-AMS inhibits saMenE with K(i)(app) =22±8 nM and ecMenE with K(i)(OSB) =128±5 nM. Putative active-site residues, Arg222, which may interact with the OSB aromatic carboxylate, and Ser302, which may bind the OSB ketone oxygen, have been identified through computational docking of OSB-AMP with the unliganded crystal structure of saMenE. A pH-dependent interconversion of the free keto acid and lactol forms of the inhibitors is also described, along with implications for inhibitor design.

FTY720 analogues as sphingosine kinase 1 inhibitors: enzyme inhibition kinetics, allosterism, proteasomal degradation, and actin rearrangement in MCF-7 breast cancer cells.

Sphingosine kinase 1 (SK1) catalyzes the conversion of sphingosine to the bioactive lipid sphingosine 1-phosphate. We have previously demonstrated that FTY720 and (S)-FTY720 vinylphosphonate are novel inhibitors of SK1 activity. Here, we show that (S)-FTY720 vinylphosphonate binds to a putative allosteric site in SK1 contingent on formation of the enzyme-sphingosine complex. We report that SK1 is an oligomeric protein (minimally a dimer) containing noncooperative catalytic sites and that the allosteric site exerts an autoinhibition of the catalytic site. A model is proposed in which (S)-FTY720 vinylphosphonate binding to and stabilization of the allosteric site might enhance the autoinhibitory effect on SK1 activity. Further evidence for the existence of allosteric site(s) in SK1 was demonstrated by data showing that two new FTY720 analogues (a conjugate of sphingosine with a fluorophore and (S)-FTY720 regioisomer) increased SK1 activity, suggesting relief of autoinhibition of SK1 activity. Comparisons with the SK1 inhibitor, SKi or siRNA knockdown of SK1 indicated that (S)-FTY720 vinylphosphonate and FTY720 behave as typical SK1 inhibitors in preventing sphingosine 1-phosphate-stimulated rearrangement of actin in MCF-7 cells. These findings are discussed in relation to the anticancer properties of SK1 inhibitors.

(S)-FTY720-vinylphosphonate, an analogue of the immunosuppressive agent FTY720, is a pan-antagonist of sphingosine 1-phosphate GPCR signaling and inhibits autotaxin activity.

FTY720 (Fingolimod), a synthetic analogue of sphingosine 1-phosphate (S1P), activates four of the five EDG-family S1P receptors and is in a phase-III clinical study for the treatment of multiple sclerosis. (S)-FTY720-phosphate (FTY720-P) causes S1P(1) receptor internalization and targeting to the proteasomal degradative pathway, and thus functions as an antagonist of S1P(1) by depleting the functional S1P(1) receptor from the plasma membrane. Here we describe the pharmacological characterization of two unsaturated phosphonate enantiomers of FTY720, (R)- and (S)-FTY720-vinylphosphonate. (R)-FTY720-vinylphosphonate was a full agonist of S1P(1) (EC(50) 20+/-3 nM). In contrast, the (S) enantiomer failed to activate any of the five S1P GPCRs and was a full antagonist of S1P(1,3,4) (K(i) 384 nM, 39 nM, and 1190 nM, respectively) and a partial antagonist of S1P(2), and S1P(5). Both enantiomers dose-dependently inhibited lysophospholipase D (recombinant autotaxin) with K(i) values in the low micromolar range, although with different enzyme kinetic mechanisms. When injected into mice, both enantiomers caused transient peripheral lymphopenia. (R)- and (S)-FTY720-vinylphosphonates activated ERK1/2, AKT, and exerted an antiapoptotic effect in camptothecin-treated IEC-6 intestinal epithelial cells, which primarily express S1P(2) transcripts and traces of S1P(5). (S)-FTY720-vinylphosphonate is the first pan-antagonist of S1P receptors and offers utility in probing S1P responses in vitro and in vivo. The biological effects of the (R)- and (S)-FTY720-vinylphosphonate analogues underscore the complexity of FTY720 cellular targets.

Active site remodelling accompanies thioester bond formation in the SUMO E1.

E1 enzymes activate ubiquitin (Ub) and ubiquitin-like (Ubl) proteins in two steps by carboxy-terminal adenylation and thioester bond formation to a conserved catalytic cysteine in the E1 Cys domain. The structural basis for these intermediates remains unknown. Here we report crystal structures for human SUMO E1 in complex with SUMO adenylate and tetrahedral intermediate analogues at 2.45 and 2.6 A, respectively. These structures show that side chain contacts to ATP.Mg are released after adenylation to facilitate a 130 degree rotation of the Cys domain during thioester bond formation that is accompanied by remodelling of key structural elements including the helix that contains the E1 catalytic cysteine, the crossover and re-entry loops, and refolding of two helices that are required for adenylation. These changes displace side chains required for adenylation with side chains required for thioester bond formation. Mutational and biochemical analyses indicate these mechanisms are conserved in other E1s.