Search

Human-Object Interaction (HOI) detection is a fundamental task in image understanding. While deep-learning-based HOI methods provide high performance in terms of mean Average Precision (mAP), they are computationally expensive and opaque in training and inference processes. An Efficient HOI (EHOI) detector is proposed in this work to strike a good balance between detection performance, inference complexity, and mathematical transparency. EHOI is a two-stage method. In the first stage, it leverages a frozen object detector to localize the objects and extract various features as intermediate outputs. In the second stage, the first-stage outputs predict the interaction type using the XGBoost classifier. Our contributions include the application of error correction codes (ECCs) to encode rare interaction cases, which reduces the model size and the complexity of the XGBoost classifier in the second stage. Additionally, we provide a mathematical formulation of the relabeling and decision-making process. Apart from the architecture, we present qualitative results to explain the functionalities of the feedforward modules. Experimental results demonstrate the advantages of ECC-coded interaction labels and the excellent balance of detection performance and complexity of the proposed EHOI method.

As a fundamental task in natural language processing, word embedding converts each word into a representation in a vector space. A challenge with word embedding is that as the vocabulary grows, the vector space's dimension increases and it can lead to a vast model size. Storing and processing word vectors are resource-demanding, especially for mobile edge-devices applications. This paper explores word embedding dimension reduction. To balance computational costs and performance, we propose an efficient and effective weakly-supervised feature selection method, named WordFS. It has two variants, each utilizing novel criteria for feature selection. Experiments conducted on various tasks (e.g., word and sentence similarity and binary and multi-class classification) indicate that the proposed WordFS model outperforms other dimension reduction methods at lower computational costs.

Blind Image Quality Assessment (BIQA) is an essential task that estimates the perceptual quality of images without reference. While many BIQA methods employ deep neural networks (DNNs) and incorporate saliency detectors to enhance performance, their large model sizes limit deployment on resource-constrained devices. To address this challenge, we introduce a novel and non-deep-learning BIQA method with a lightweight saliency detection module, called Green Saliency-guided Blind Image Quality Assessment (GSBIQA). It is characterized by its minimal model size, reduced computational demands, and robust performance. Experimental results show that the performance of GSBIQA is comparable with state-of-the-art DL-based methods with significantly lower resource requirements.

We introduce GreenCOD, a green method for detecting camouflaged objects, distinct in its avoidance of backpropagation techniques. GreenCOD leverages gradient boosting and deep features extracted from pre-trained Deep Neural Networks (DNNs). Traditional camouflaged object detection (COD) approaches often rely on complex deep neural network architectures, seeking performance improvements through backpropagation-based fine-tuning. However, such methods are typically computationally demanding and exhibit only marginal performance variations across different models. This raises the question of whether effective training can be achieved without backpropagation. Addressing this, our work proposes a new paradigm that utilizes gradient boosting for COD. This approach significantly simplifies the model design, resulting in a system that requires fewer parameters and operations and maintains high performance compared to state-of-the-art deep learning models. Remarkably, our models are trained without backpropagation and achieve the best performance with fewer than 20G Multiply-Accumulate Operations (MACs). This new, more efficient paradigm opens avenues for further exploration in green, backpropagation-free model training.

Image saliency detection is crucial in understanding human gaze patterns from visual stimuli. The escalating demand for research in image saliency detection is driven by the growing necessity to incorporate such techniques into various computer vision tasks and to understand human visual systems. Many existing image saliency detection methods rely on deep neural networks (DNNs) to achieve good performance. However, the high computational complexity associated with these approaches impedes their integration with other modules or deployment on resource-constrained platforms, such as mobile devices. To address this need, we propose a novel image saliency detection method named GreenSaliency, which has a small model size, minimal carbon footprint, and low computational complexity. GreenSaliency can be a competitive alternative to the existing deep-learning-based (DL-based) image saliency detection methods with limited computation resources. GreenSaliency comprises two primary steps: 1) multi-layer hybrid feature extraction and 2) multi-path saliency prediction. Experimental results demonstrate that GreenSaliency achieves comparable performance to the state-of-the-art DL-based methods while possessing a considerably smaller model size and significantly reduced computational complexity.

Automatic prostate segmentation is an important step in computer-aided diagnosis of prostate cancer and treatment planning. Existing methods of prostate segmentation are based on deep learning models which have a large size and lack of transparency which is essential for physicians. In this paper, a new data-driven 3D prostate segmentation method on MRI is proposed, named PSHop. Different from deep learning based methods, the core methodology of PSHop is a feed-forward encoder-decoder system based on successive subspace learning (SSL). It consists of two modules: 1) encoder: fine to coarse unsupervised representation learning with cascaded VoxelHop units, 2) decoder: coarse to fine segmentation prediction with voxel-wise classification and local refinement. Experiments are conducted on the publicly available ISBI-2013 dataset, as well as on a larger private one. Experimental analysis shows that our proposed PSHop is effective, robust and lightweight in the tasks of prostate gland and zonal segmentation, achieving a Dice Similarity Coefficient (DSC) of 0.873 for the gland segmentation task. PSHop achieves a competitive performance comparatively to other deep learning methods, while keeping the model size and inference complexity an order of magnitude smaller., Comment: 11 pages, 5 figures, 5 tables

Prostate Cancer is one of the most frequently occurring cancers in men, with a low survival rate if not early diagnosed. PI-RADS reading has a high false positive rate, thus increasing the diagnostic incurred costs and patient discomfort. Deep learning (DL) models achieve a high segmentation performance, although require a large model size and complexity. Also, DL models lack of feature interpretability and are perceived as ``black-boxes" in the medical field. PCa-RadHop pipeline is proposed in this work, aiming to provide a more transparent feature extraction process using a linear model. It adopts the recently introduced Green Learning (GL) paradigm, which offers a small model size and low complexity. PCa-RadHop consists of two stages: Stage-1 extracts data-driven radiomics features from the bi-parametric Magnetic Resonance Imaging (bp-MRI) input and predicts an initial heatmap. To reduce the false positive rate, a subsequent stage-2 is introduced to refine the predictions by including more contextual information and radiomics features from each already detected Region of Interest (ROI). Experiments on the largest publicly available dataset, PI-CAI, show a competitive performance standing of the proposed method among other deep DL models, achieving an area under the curve (AUC) of 0.807 among a cohort of 1,000 patients. Moreover, PCa-RadHop maintains orders of magnitude smaller model size and complexity., Comment: 13 pages, 7 figures, 5 tables

In this work, we aim to predict the survival time (ST) of glioblastoma (GBM) patients undergoing different treatments based on preoperative magnetic resonance (MR) scans. The personalized and precise treatment planning can be achieved by comparing the ST of different treatments. It is well established that both the current status of the patient (as represented by the MR scans) and the choice of treatment are the cause of ST. While previous related MR-based glioblastoma ST studies have focused only on the direct mapping of MR scans to ST, they have not included the underlying causal relationship between treatments and ST. To address this limitation, we propose a treatment-conditioned regression model for glioblastoma ST that incorporates treatment information in addition to MR scans. Our approach allows us to effectively utilize the data from all of the treatments in a unified manner, rather than having to train separate models for each of the treatments. Furthermore, treatment can be effectively injected into each convolutional layer through the adaptive instance normalization we employ. We evaluate our framework on the BraTS20 ST prediction task. Three treatment options are considered: Gross Total Resection (GTR), Subtotal Resection (STR), and no resection. The evaluation results demonstrate the effectiveness of injecting the treatment for estimating GBM survival., Comment: SPIE Medical Imaging 2024: Computer-Aided Diagnosis

AI algorithms at the edge demand smaller model sizes and lower computational complexity. To achieve these objectives, we adopt a green learning (GL) paradigm rather than the deep learning paradigm. GL has three modules: 1) unsupervised representation learning, 2) supervised feature learning, and 3) supervised decision learning. We focus on the second module in this work. In particular, we derive new discriminant features from proper linear combinations of input features, denoted by x, obtained in the first module. They are called complementary and raw features, respectively. Along this line, we present a novel supervised learning method to generate highly discriminant complementary features based on the least-squares normal transform (LNT). LNT consists of two steps. First, we convert a C-class classification problem to a binary classification problem. The two classes are assigned with 0 and 1, respectively. Next, we formulate a least-squares regression problem from the N-dimensional (N-D) feature space to the 1-D output space, and solve the least-squares normal equation to obtain one N-D normal vector, denoted by a1. Since one normal vector is yielded by one binary split, we can obtain M normal vectors with M splits. Then, Ax is called an LNT of x, where transform matrix A in R^{M by N} by stacking aj^T, j=1, ..., M, and the LNT, Ax, can generate M new features. The newly generated complementary features are shown to be more discriminant than the raw features. Experiments show that the classification performance can be improved by these new features., Comment: 2023 IEEE International Conference on Big Data, AI and Adaptive Computing for Edge Sensing and Processing Workshop

Unsupervised image-to-image (I2I) translation learns cross-domain image mapping that transfers input from the source domain to output in the target domain while preserving its semantics. One challenge is that different semantic statistics in source and target domains result in content discrepancy known as semantic distortion. To address this problem, a novel I2I method that maintains semantic consistency in translation is proposed and named SemST in this work. SemST reduces semantic distortion by employing contrastive learning and aligning the structural and textural properties of input and output by maximizing their mutual information. Furthermore, a multi-scale approach is introduced to enhance translation performance, thereby enabling the applicability of SemST to domain adaptation in high-resolution images. Experiments show that SemST effectively mitigates semantic distortion and achieves state-of-the-art performance. Also, the application of SemST to domain adaptation (DA) is explored. It is demonstrated by preliminary experiments that SemST can be utilized as a beneficial pre-training for the semantic segmentation task.

Many mathematical models have been leveraged to design embeddings for representing Knowledge Graph (KG) entities and relations for link prediction and many downstream tasks. These mathematically-inspired models are not only highly scalable for inference in large KGs, but also have many explainable advantages in modeling different relation patterns that can be validated through both formal proofs and empirical results. In this paper, we make a comprehensive overview of the current state of research in KG completion. In particular, we focus on two main branches of KG embedding (KGE) design: 1) distance-based methods and 2) semantic matching-based methods. We discover the connections between recently proposed models and present an underlying trend that might help researchers invent novel and more effective models. Next, we delve into CompoundE and CompoundE3D, which draw inspiration from 2D and 3D affine operations, respectively. They encompass a broad spectrum of techniques including distance-based and semantic-based methods. We will also discuss an emerging approach for KG completion which leverages pre-trained language models (PLMs) and textual descriptions of entities and relations and offer insights into the integration of KGE embedding methods with PLMs for KG completion.

Supervised trackers trained on labeled data dominate the single object tracking field for superior tracking accuracy. The labeling cost and the huge computational complexity hinder their applications on edge devices. Unsupervised learning methods have also been investigated to reduce the labeling cost but their complexity remains high. Aiming at lightweight high-performance tracking, feasibility without offline pre-training, and algorithmic transparency, we propose a new single object tracking method, called the green object tracker (GOT), in this work. GOT conducts an ensemble of three prediction branches for robust box tracking: 1) a global object-based correlator to predict the object location roughly, 2) a local patch-based correlator to build temporal correlations of small spatial units, and 3) a superpixel-based segmentator to exploit the spatial information of the target frame. GOT offers competitive tracking accuracy with state-of-the-art unsupervised trackers, which demand heavy offline pre-training, at a lower computation cost. GOT has a tiny model size (<3k parameters) and low inference complexity (around 58M FLOPs per frame). Since its inference complexity is between 0.1%-10% of DL trackers, it can be easily deployed on mobile and edge devices.

Chatbots have been studied for more than half a century. With the rapid development of natural language processing (NLP) technologies in recent years, chatbots using large language models (LLMs) have received much attention nowadays. Compared with traditional ones, modern chatbots are more powerful and have been used in real-world applications. There are however, bias and fairness concerns in modern chatbot design. Due to the huge amounts of training data, extremely large model sizes, and lack of interpretability, bias mitigation and fairness preservation of modern chatbots are challenging. Thus, a comprehensive overview on bias and fairness in chatbot systems is given in this paper. The history of chatbots and their categories are first reviewed. Then, bias sources and potential harms in applications are analyzed. Considerations in designing fair and unbiased chatbot systems are examined. Finally, future research directions are discussed.

Knowledge graph entity typing (KGET) is a task to predict the missing entity types in knowledge graphs (KG). Previously, KG embedding (KGE) methods tried to solve the KGET task by introducing an auxiliary relation, 'hasType', to model the relationship between entities and their types. However, a single auxiliary relation has limited expressiveness for diverse entity-type patterns. We improve the expressiveness of KGE methods by introducing multiple auxiliary relations in this work. Similar entity types are grouped to reduce the number of auxiliary relations and improve their capability to model entity-type patterns with different granularities. With the presence of multiple auxiliary relations, we propose a method adopting an Asynchronous learning scheme for Entity Typing, named AsyncET, which updates the entity and type embeddings alternatively to keep the learned entity embedding up-to-date and informative for entity type prediction. Experiments are conducted on two commonly used KGET datasets to show that the performance of KGE methods on the KGET task can be substantially improved by the proposed multiple auxiliary relations and asynchronous embedding learning. Furthermore, our method has a significant advantage over state-of-the-art methods in model sizes and time complexity.

In the cancer diagnosis pipeline, digital pathology plays an instrumental role in the identification, staging, and grading of malignant areas on biopsy tissue specimens. High resolution histology images are subject to high variance in appearance, sourcing either from the acquisition devices or the H\&E staining process. Nuclei segmentation is an important task, as it detects the nuclei cells over background tissue and gives rise to the topology, size, and count of nuclei which are determinant factors for cancer detection. Yet, it is a fairly time consuming task for pathologists, with reportedly high subjectivity. Computer Aided Diagnosis (CAD) tools empowered by modern Artificial Intelligence (AI) models enable the automation of nuclei segmentation. This can reduce the subjectivity in analysis and reading time. This paper provides an extensive review, beginning from earlier works use traditional image processing techniques and reaching up to modern approaches following the Deep Learning (DL) paradigm. Our review also focuses on the weak supervision aspect of the problem, motivated by the fact that annotated data is scarce. At the end, the advantages of different models and types of supervision are thoroughly discussed. Furthermore, we try to extrapolate and envision how future research lines will potentially be, so as to minimize the need for labeled data while maintaining high performance. Future methods should emphasize efficient and explainable models with a transparent underlying process so that physicians can trust their output., Comment: 47 pages, 27 figures, 9 tables

Owing to the proliferation of user-generated videos on the Internet, blind video quality assessment (BVQA) at the edge attracts growing attention. The usage of deep-learning-based methods is restricted to be applied at the edge due to their large model sizes and high computational complexity. In light of this, a novel lightweight BVQA method called GreenBVQA is proposed in this work. GreenBVQA features a small model size, low computational complexity, and high performance. Its processing pipeline includes: video data cropping, unsupervised representation generation, supervised feature selection, and mean-opinion-score (MOS) regression and ensembles. We conduct experimental evaluations on three BVQA datasets and show that GreenBVQA can offer state-of-the-art performance in PLCC and SROCC metrics while demanding significantly smaller model sizes and lower computational complexity. Thus, GreenBVQA is well-suited for edge devices.

A novel learning solution to image steganalysis based on the green learning paradigm, called Green Steganalyzer (GS), is proposed in this work. GS consists of three modules: 1) pixel-based anomaly prediction, 2) embedding location detection, and 3) decision fusion for image-level detection. In the first module, GS decomposes an image into patches, adopts Saab transforms for feature extraction, and conducts self-supervised learning to predict an anomaly score of their center pixel. In the second module, GS analyzes the anomaly scores of a pixel and its neighborhood to find pixels of higher embedding probabilities. In the third module, GS focuses on pixels of higher embedding probabilities and fuses their anomaly scores to make final image-level classification. Compared with state-of-the-art deep-learning models, GS achieves comparable detection performance against S-UNIWARD, WOW and HILL steganography schemes with significantly lower computational complexity and a smaller model size, making it attractive for mobile/edge applications. Furthermore, GS is mathematically transparent because of its modular design.

As a specific category of artificial intelligence (AI), generative artificial intelligence (GenAI) generates new content that resembles what is created by humans. The rapid development of GenAI systems has created a huge amount of new data on the Internet, posing new challenges to current computing and communication frameworks. Currently, GenAI services rely on the traditional cloud computing framework due to the need for large computation resources. However, such services will encounter high latency because of data transmission and a high volume of requests. On the other hand, edge-cloud computing can provide adequate computation power and low latency at the same time through the collaboration between edges and the cloud. Thus, it is attractive to build GenAI systems at scale by leveraging the edge-cloud computing paradigm. In this overview paper, we review recent developments in GenAI and edge-cloud computing, respectively. Then, we use two exemplary GenAI applications to discuss technical challenges in scaling up their solutions using edge-cloud collaborative systems. Finally, we list design considerations for training and deploying GenAI systems at scale and point out future research directions.

Ensuring the realism of computer-generated synthetic images is crucial to deep neural network (DNN) training. Due to different semantic distributions between synthetic and real-world captured datasets, there exists semantic mismatch between synthetic and refined images, which in turn results in the semantic distortion. Recently, contrastive learning (CL) has been successfully used to pull correlated patches together and push uncorrelated ones apart. In this work, we exploit semantic and structural consistency between synthetic and refined images and adopt CL to reduce the semantic distortion. Besides, we incorporate hard negative mining to improve the performance furthermore. We compare the performance of our method with several other benchmarking methods using qualitative and quantitative measures and show that our method offers the state-of-the-art performance.

The cascade of 2D geometric transformations were exploited to model relations between entities in a knowledge graph (KG), leading to an effective KG embedding (KGE) model, CompoundE. Furthermore, the rotation in the 3D space was proposed as a new KGE model, Rotate3D, by leveraging its non-commutative property. Inspired by CompoundE and Rotate3D, we leverage 3D compound geometric transformations, including translation, rotation, scaling, reflection, and shear and propose a family of KGE models, named CompoundE3D, in this work. CompoundE3D allows multiple design variants to match rich underlying characteristics of a KG. Since each variant has its own advantages on a subset of relations, an ensemble of multiple variants can yield superior performance. The effectiveness and flexibility of CompoundE3D are experimentally verified on four popular link prediction datasets.

Blind image quality assessment (BIQA) is a task that predicts the perceptual quality of an image without its reference. Research on BIQA attracts growing attention due to the increasing amount of user-generated images and emerging mobile applications where reference images are unavailable. The problem is challenging due to the wide range of content and mixed distortion types. Many existing BIQA methods use deep neural networks (DNNs) to achieve high performance. However, their large model sizes hinder their applicability to edge or mobile devices. To meet the need, a novel BIQA method with a small model, low computational complexity, and high performance is proposed and named "GreenBIQA" in this work. GreenBIQA includes five steps: 1) image cropping, 2) unsupervised representation generation, 3) supervised feature selection, 4) distortion-specific prediction, and 5) regression and decision ensemble. Experimental results show that the performance of GreenBIQA is comparable with that of state-of-the-art deep-learning (DL) solutions while demanding a much smaller model size and significantly lower computational complexity.

The design of a tiny machine learning model, which can be deployed in mobile and edge devices, for point cloud object classification is investigated in this work. To achieve this objective, we replace the multi-scale representation of a point cloud object with a single-scale representation for complexity reduction, and exploit rich 3D geometric information of a point cloud object for performance improvement. The proposed solution is named Green-PointHop due to its low computational complexity. We evaluate the performance of Green-PointHop on ModelNet40 and ScanObjectNN two datasets. Green-PointHop has a model size of 64K parameters. It demands 2.3M floating-point operations (FLOPs) to classify a ModelNet40 object of 1024 down-sampled points. Its classification performance gaps against the state-of-the-art DGCNN method are 3% and 7% for ModelNet40 and ScanObjectNN, respectively. On the other hand, the model size and inference complexity of DGCNN are 42X and 1203X of those of Green-PointHop, respectively., Comment: 13 pages, 4 figures

Language modeling studies the probability distributions over strings of texts. It is one of the most fundamental tasks in natural language processing (NLP). It has been widely used in text generation, speech recognition, machine translation, etc. Conventional language models (CLMs) aim to predict the probability of linguistic sequences in a causal manner, while pre-trained language models (PLMs) cover broader concepts and can be used in both causal sequential modeling and fine-tuning for downstream applications. PLMs have their own training paradigms (usually self-supervised) and serve as foundation models in modern NLP systems. This overview paper provides an introduction to both CLMs and PLMs from five aspects, i.e., linguistic units, architectures, training methods, evaluation methods, and applications. Furthermore, we discuss the relationship between CLMs and PLMs and shed light on the future directions of language modeling in the pre-trained era.

A light-weight super-resolution (LSR) method from a single image targeting mobile applications is proposed in this work. LSR predicts the residual image between the interpolated low-resolution (ILR) and high-resolution (HR) images using a self-supervised framework. To lower the computational complexity, LSR does not adopt the end-to-end optimization deep networks. It consists of three modules: 1) generation of a pool of rich and diversified representations in the neighborhood of a target pixel via unsupervised learning, 2) selecting a subset from the representation pool that is most relevant to the underlying super-resolution task automatically via supervised learning, 3) predicting the residual of the target pixel via regression. LSR has low computational complexity and reasonable model size so that it can be implemented on mobile/edge platforms conveniently. Besides, it offers better visual quality than classical exemplar-based methods in terms of PSNR/SSIM measures., Comment: 8 pages, 3 figures, 10 tables

An efficient 3D scene flow estimation method called PointFlowHop is proposed in this work. PointFlowHop takes two consecutive point clouds and determines the 3D flow vectors for every point in the first point cloud. PointFlowHop decomposes the scene flow estimation task into a set of subtasks, including ego-motion compensation, object association and object-wise motion estimation. It follows the green learning (GL) pipeline and adopts the feedforward data processing path. As a result, its underlying mechanism is more transparent than deep-learning (DL) solutions based on end-to-end optimization of network parameters. We conduct experiments on the stereoKITTI and the Argoverse LiDAR point cloud datasets and demonstrate that PointFlowHop outperforms deep-learning methods with a small model size and less training time. Furthermore, we compare the Floating Point Operations (FLOPs) required by PointFlowHop and other learning-based methods in inference, and show its big savings in computational complexity., Comment: 13 pages, 5 figures

Many point cloud classification methods are developed under the assumption that all point clouds in the dataset are well aligned with the canonical axes so that the 3D Cartesian point coordinates can be employed to learn features. When input point clouds are not aligned, the classification performance drops significantly. In this work, we focus on a mathematically transparent point cloud classification method called PointHop, analyze its reason for failure due to pose variations, and solve the problem by replacing its pose dependent modules with rotation invariant counterparts. The proposed method is named SO(3)-Invariant PointHop (or S3I-PointHop in short). We also significantly simplify the PointHop pipeline using only one single hop along with multiple spatial aggregation techniques. The idea of exploiting more spatial information is novel. Experiments on the ModelNet40 dataset demonstrate the superiority of S3I-PointHop over traditional PointHop-like methods., Comment: 5 pages, 3 figures

A low-complexity point cloud compression method called the Green Point Cloud Geometry Codec (GPCGC), is proposed to encode the 3D spatial coordinates of static point clouds efficiently. GPCGC consists of two modules. In the first module, point coordinates of input point clouds are hierarchically organized into an octree structure. Points at each leaf node are projected along one of three axes to yield image maps. In the second module, the occupancy map is clustered into 9 modes while the depth map is coded by a low-complexity high-efficiency image codec, called the green image codec (GIC). GIC is a multi-resolution codec based on vector quantization (VQ). Its complexity is significantly lower than HEVC-Intra. Furthermore, the rate-distortion optimization (RDO) technique is used to select the optimal coding parameters. GPCGC is a progressive codec, and it offers a coding performance competitive with MPEG's V-PCC and G-PCC standards at significantly lower complexity.

Cardiac cine magnetic resonance imaging (MRI) has been used to characterize cardiovascular diseases (CVD), often providing a noninvasive phenotyping tool.~While recently flourished deep learning based approaches using cine MRI yield accurate characterization results, the performance is often degraded by small training samples. In addition, many deep learning models are deemed a ``black box," for which models remain largely elusive in how models yield a prediction and how reliable they are. To alleviate this, this work proposes a lightweight successive subspace learning (SSL) framework for CVD classification, based on an interpretable feedforward design, in conjunction with a cardiac atlas. Specifically, our hierarchical SSL model is based on (i) neighborhood voxel expansion, (ii) unsupervised subspace approximation, (iii) supervised regression, and (iv) multi-level feature integration. In addition, using two-phase 3D deformation fields, including end-diastolic and end-systolic phases, derived between the atlas and individual subjects as input offers objective means of assessing CVD, even with small training samples. We evaluate our framework on the ACDC2017 database, comprising one healthy group and four disease groups. Compared with 3D CNN-based approaches, our framework achieves superior classification performance with 140$\times$ fewer parameters, which supports its potential value in clinical use., Comment: ISBI 2023

A self-supervised adaptive low-light video enhancement method, called SALVE, is proposed in this work. SALVE first enhances a few key frames of an input low-light video using a retinex-based low-light image enhancement technique. For each keyframe, it learns a mapping from low-light image patches to enhanced ones via ridge regression. These mappings are then used to enhance the remaining frames in the low-light video. The combination of traditional retinex-based image enhancement and learning-based ridge regression leads to a robust, adaptive and computationally inexpensive solution to enhance low-light videos. Our extensive experiments along with a user study show that 87% of participants prefer SALVE over prior work., Comment: 12 pages, 7 figures, 4 tables

Despite prolific work on evaluating generative models, little research has been done on the quality evaluation of an individual generated sample. To address this problem, a lightweight generated sample quality evaluation (LGSQE) method is proposed in this work. In the training stage of LGSQE, a binary classifier is trained on real and synthetic samples, where real and synthetic data are labeled by 0 and 1, respectively. In the inference stage, the classifier assigns soft labels (ranging from 0 to 1) to each generated sample. The value of soft label indicates the quality level; namely, the quality is better if its soft label is closer to 0. LGSQE can serve as a post-processing module for quality control. Furthermore, LGSQE can be used to evaluate the performance of generative models, such as accuracy, AUC, precision and recall, by aggregating sample-level quality. Experiments are conducted on CIFAR-10 and MNIST to demonstrate that LGSQE can preserve the same performance rank order as that predicted by the Frechet Inception Distance (FID) but with significantly lower complexity.

Recovering unknown, missing, damaged, distorted, or lost information in DCT coefficients is a common task in multiple applications of digital image processing, including image compression, selective image encryption, and image communication. This paper investigates the recovery of sign bits in DCT coefficients of digital images, by proposing two different approximation methods to solve a mixed integer linear programming (MILP) problem, which is NP-hard in general. One method is a relaxation of the MILP problem to a linear programming (LP) problem, and the other splits the original MILP problem into some smaller MILP problems and an LP problem. We considered how the proposed methods can be applied to JPEG-encoded images and conducted extensive experiments to validate their performances. The experimental results showed that the proposed methods outperformed other existing methods by a substantial margin, both according to objective quality metrics and our subjective evaluation., Comment: 22 pages, 8 figures

Being different from deep-learning-based (DL-based) image generation methods, a new image generative model built upon successive subspace learning principle is proposed and named GenHop (an acronym of Generative PixelHop) in this work. GenHop consists of three modules: 1) high-to-low dimension reduction, 2) seed image generation, and 3) low-to-high dimension expansion. In the first module, it builds a sequence of high-to-low dimensional subspaces through a sequence of whitening processes, each of which contains samples of joint-spatial-spectral representation. In the second module, it generates samples in the lowest dimensional subspace. In the third module, it finds a proper high-dimensional sample for a seed image by adding details back via locally linear embedding (LLE) and a sequence of coloring processes. Experiments show that GenHop can generate visually pleasant images whose FID scores are comparable or even better than those of DL-based generative models for MNIST, Fashion-MNIST and CelebA datasets., Comment: 10 pages, 5 figures, accepted by ISCAS 2022

Rapid advances in artificial intelligence (AI) in the last decade have largely been built upon the wide applications of deep learning (DL). However, the high carbon footprint yielded by larger and larger DL networks becomes a concern for sustainability. Furthermore, DL decision mechanism is somewhat obsecure and can only be verified by test data. Green learning (GL) has been proposed as an alternative paradigm to address these concerns. GL is characterized by low carbon footprints, small model sizes, low computational complexity, and logical transparency. It offers energy-effective solutions in cloud centers as well as mobile/edge devices. GL also provides a clear and logical decision-making process to gain people's trust. Several statistical tools have been developed to achieve this goal in recent years. They include subspace approximation, unsupervised and supervised representation learning, supervised discriminant feature selection, and feature space partitioning. We have seen a few successful GL examples with performance comparable with state-of-the-art DL solutions. This paper offers an introduction to GL, its demonstrated applications, and future outlook.

A multi-grid multi-block-size vector quantization (MGBVQ) method is proposed for image coding in this work. The fundamental idea of image coding is to remove correlations among pixels before quantization and entropy coding, e.g., the discrete cosine transform (DCT) and intra predictions, adopted by modern image coding standards. We present a new method to remove pixel correlations. First, by decomposing correlations into long- and short-range correlations, we represent long-range correlations in coarser grids due to their smoothness, thus leading to a multi-grid (MG) coding architecture. Second, we show that short-range correlations can be effectively coded by a suite of vector quantizers (VQs). Along this line, we argue the effectiveness of VQs of very large block sizes and present a convenient way to implement them. It is shown by experimental results that MGBVQ offers excellent rate-distortion (RD) performance, which is comparable with existing image coders, at much lower complexity. Besides, it provides a progressive coded bitstream., Comment: GIC-python-v2

We offer a method for one-shot mask-guided image synthesis that allows controlling manipulations of a single image by inverting a quasi-robust classifier equipped with strong regularizers. Our proposed method, entitled MAGIC, leverages structured gradients from a pre-trained quasi-robust classifier to better preserve the input semantics while preserving its classification accuracy, thereby guaranteeing credibility in the synthesis. Unlike current methods that use complex primitives to supervise the process or use attention maps as a weak supervisory signal, MAGIC aggregates gradients over the input, driven by a guide binary mask that enforces a strong, spatial prior. MAGIC implements a series of manipulations with a single framework achieving shape and location control, intense non-rigid shape deformations, and copy/move operations in the presence of repeating objects and gives users firm control over the synthesis by requiring to simply specify binary guide masks. Our study and findings are supported by various qualitative comparisons with the state-of-the-art on the same images sampled from ImageNet and quantitative analysis using machine perception along with a user survey of 100+ participants that endorse our synthesis quality. Project page at https://mozhdehrouhsedaghat.github.io/magic.html. Code is available at https://github.com/mozhdehrouhsedaghat/magic, Comment: Accepted to the Thirty-Seventh Conference on Artificial Intelligence (AAAI) 2023 - 12 pages, 9 figures

Knowledge graph completion (KGC) aims to discover missing relationships between entities in knowledge graphs (KGs). Most prior KGC work focuses on learning embeddings for entities and relations through a simple scoring function. Yet, a higher-dimensional embedding space is usually required for a better reasoning capability, which leads to a larger model size and hinders applicability to real-world problems (e.g., large-scale KGs or mobile/edge computing). A lightweight modularized KGC solution, called GreenKGC, is proposed in this work to address this issue. GreenKGC consists of three modules: representation learning, feature pruning, and decision learning, to extract discriminant KG features and make accurate predictions on missing relationships using classifiers and negative sampling. Experimental results demonstrate that, in low dimensions, GreenKGC can outperform SOTA methods in most datasets. In addition, low-dimensional GreenKGC can achieve competitive or even better performance against high-dimensional models with a much smaller model size., Comment: Accepted to ACL2023

Unsupervised domain adaptation (UDA) has been widely used to transfer knowledge from a labeled source domain to an unlabeled target domain to counter the difficulty of labeling in a new domain. The training of conventional solutions usually relies on the existence of both source and target domain data. However, privacy of the large-scale and well-labeled data in the source domain and trained model parameters can become the major concern of cross center/domain collaborations. In this work, to address this, we propose a practical solution to UDA for segmentation with a black-box segmentation model trained in the source domain only, rather than original source data or a white-box source model. Specifically, we resort to a knowledge distillation scheme with exponential mixup decay (EMD) to gradually learn target-specific representations. In addition, unsupervised entropy minimization is further applied to regularization of the target domain confidence. We evaluated our framework on the BraTS 2018 database, achieving performance on par with white-box source model adaptation approaches., Comment: SPIE Medical Imaging 2022: Image Processing

Unsupervised domain adaptation (UDA) has been successfully applied to transfer knowledge from a labeled source domain to target domains without their labels. Recently introduced transferable prototypical networks (TPN) further addresses class-wise conditional alignment. In TPN, while the closeness of class centers between source and target domains is explicitly enforced in a latent space, the underlying fine-grained subtype structure and the cross-domain within-class compactness have not been fully investigated. To counter this, we propose a new approach to adaptively perform a fine-grained subtype-aware alignment to improve performance in the target domain without the subtype label in both domains. The insight of our approach is that the unlabeled subtypes in a class have the local proximity within a subtype, while exhibiting disparate characteristics, because of different conditional and label shifts. Specifically, we propose to simultaneously enforce subtype-wise compactness and class-wise separation, by utilizing intermediate pseudo-labels. In addition, we systematically investigate various scenarios with and without prior knowledge of subtype numbers, and propose to exploit the underlying subtype structure. Furthermore, a dynamic queue framework is developed to evolve the subtype cluster centroids steadily using an alternative processing scheme. Experimental results, carried out with multi-view congenital heart disease data and VisDA and DomainNet, show the effectiveness and validity of our subtype-aware UDA, compared with state-of-the-art UDA methods., Comment: IEEE Transactions on Neural Networks and Learning Systems (TNNLS)

Built upon the decision tree (DT) classification and regression idea, the subspace learning machine (SLM) has been recently proposed to offer higher performance in general classification and regression tasks. Its performance improvement is reached at the expense of higher computational complexity. In this work, we investigate two ways to accelerate SLM. First, we adopt the particle swarm optimization (PSO) algorithm to speed up the search of a discriminant dimension that is expressed as a linear combination of current dimensions. The search of optimal weights in the linear combination is computationally heavy. It is accomplished by probabilistic search in original SLM. The acceleration of SLM by PSO requires 10-20 times fewer iterations. Second, we leverage parallel processing in the SLM implementation. Experimental results show that the accelerated SLM method achieves a speed up factor of 577 in training time while maintaining comparable classification/regression performance of original SLM.

Human-centered AI considers human experiences with AI performance. While abundant research has been helping AI achieve superhuman performance either by fully automatic or weak supervision learning, fewer endeavors are experimenting with how AI can tailor to humans' preferred skill level given fine-grained input. In this work, we guide the curriculum reinforcement learning results towards a preferred performance level that is neither too hard nor too easy via learning from the human decision process. To achieve this, we developed a portable, interactive platform that enables the user to interact with agents online via manipulating the task difficulty, observing performance, and providing curriculum feedback. Our system is highly parallelizable, making it possible for a human to train large-scale reinforcement learning applications that require millions of samples without a server. The result demonstrates the effectiveness of an interactive curriculum for reinforcement learning involving human-in-the-loop. It shows reinforcement learning performance can successfully adjust in sync with the human desired difficulty level. We believe this research will open new doors for achieving flow and personalized adaptive difficulties., Comment: 6 pages, 7 figures

A statistical attention localization (SAL) method is proposed to facilitate the object classification task in this work. SAL consists of three steps: 1) preliminary attention window selection via decision statistics, 2) attention map refinement, and 3) rectangular attention region finalization. SAL computes soft-decision scores of local squared windows and uses them to identify salient regions in Step 1. To accommodate object of various sizes and shapes, SAL refines the preliminary result and obtain an attention map of more flexible shape in Step 2. Finally, SAL yields a rectangular attention region using the refined attention map and bounding box regularization in Step 3. As an application, we adopt E-PixelHop, which is an object classification solution based on successive subspace learning (SSL), as the baseline. We apply SAL so as to obtain a cropped-out and resized attention region as an alternative input. Classification results of the whole image as well as the attention region are ensembled to achieve the highest classification accuracy. Experiments on the CIFAR-10 dataset are given to demonstrate the advantage of the SAL-assisted object classification method., Comment: 11 pages, 9 figures

Language modality within the vision language pretraining framework is innately discretized, endowing each word in the language vocabulary a semantic meaning. In contrast, visual modality is inherently continuous and high-dimensional, which potentially prohibits the alignment as well as fusion between vision and language modalities. We therefore propose to "discretize" the visual representation by joint learning a codebook that imbues each visual token a semantic. We then utilize these discretized visual semantics as self-supervised ground-truths for building our Masked Image Modeling objective, a counterpart of Masked Language Modeling which proves successful for language models. To optimize the codebook, we extend the formulation of VQ-VAE which gives a theoretic guarantee. Experiments validate the effectiveness of our approach across common vision-language benchmarks., Comment: 7 pages, 4 figures, ICPR2022. arXiv admin note: text overlap with arXiv:2203.00048

Normalizing flow models have been used successfully for generative image super-resolution (SR) by approximating complex distribution of natural images to simple tractable distribution in latent space through Invertible Neural Networks (INN). These models can generate multiple realistic SR images from one low-resolution (LR) input using randomly sampled points in the latent space, simulating the ill-posed nature of image upscaling where multiple high-resolution (HR) images correspond to the same LR. Lately, the invertible process in INN has also been used successfully by bidirectional image rescaling models like IRN and HCFlow for joint optimization of downscaling and inverse upscaling, resulting in significant improvements in upscaled image quality. While they are optimized for image downscaling too, the ill-posed nature of image downscaling, where one HR image could be downsized to multiple LR images depending on different interpolation kernels and resampling methods, is not considered. A new downscaling latent variable, in addition to the original one representing uncertainties in image upscaling, is introduced to model variations in the image downscaling process. This dual latent variable enhancement is applicable to different image rescaling models and it is shown in extensive experiments that it can improve image upscaling accuracy consistently without sacrificing image quality in downscaled LR images. It is also shown to be effective in enhancing other INN-based models for image restoration applications like image hiding., Comment: Accepted by ACM Multimedia 2022

Supervised and unsupervised deep trackers that rely on deep learning technologies are popular in recent years. Yet, they demand high computational complexity and a high memory cost. A green unsupervised single-object tracker, called GUSOT, that aims at object tracking for long videos under a resource-constrained environment is proposed in this work. Built upon a baseline tracker, UHP-SOT++, which works well for short-term tracking, GUSOT contains two additional new modules: 1) lost object recovery, and 2) color-saliency-based shape proposal. They help resolve the tracking loss problem and offer a more flexible object proposal, respectively. Thus, they enable GUSOT to achieve higher tracking accuracy in the long run. We conduct experiments on the large-scale dataset LaSOT with long video sequences, and show that GUSOT offers a lightweight high-performance tracking solution that finds applications in mobile and edge computing platforms.

Translation, rotation, and scaling are three commonly used geometric manipulation operations in image processing. Besides, some of them are successfully used in developing effective knowledge graph embedding (KGE) models such as TransE and RotatE. Inspired by the synergy, we propose a new KGE model by leveraging all three operations in this work. Since translation, rotation, and scaling operations are cascaded to form a compound one, the new model is named CompoundE. By casting CompoundE in the framework of group theory, we show that quite a few scoring-function-based KGE models are special cases of CompoundE. CompoundE extends the simple distance-based relation to relation-dependent compound operations on head and/or tail entities. To demonstrate the effectiveness of CompoundE, we conduct experiments on three popular KG completion datasets. Experimental results show that CompoundE consistently achieves the state of-the-art performance., Comment: 16 pages

Deep neural networks (DNNs) achieve great success in blind image quality assessment (BIQA) with large pre-trained models in recent years. Their solutions cannot be easily deployed at mobile or edge devices, and a lightweight solution is desired. In this work, we propose a novel BIQA model, called GreenBIQA, that aims at high performance, low computational complexity and a small model size. GreenBIQA adopts an unsupervised feature generation method and a supervised feature selection method to extract quality-aware features. Then, it trains an XGBoost regressor to predict quality scores of test images. We conduct experiments on four popular IQA datasets, which include two synthetic-distortion and two authentic-distortion datasets. Experimental results show that GreenBIQA is competitive in performance against state-of-the-art DNNs with lower complexity and smaller model sizes.

Word Mover's Distance (WMD) computes the distance between words and models text similarity with the moving cost between words in two text sequences. Yet, it does not offer good performance in sentence similarity evaluation since it does not incorporate word importance and fails to take inherent contextual and structural information in a sentence into account. An improved WMD method using the syntactic parse tree, called Syntax-aware Word Mover's Distance (SynWMD), is proposed to address these two shortcomings in this work. First, a weighted graph is built upon the word co-occurrence statistics extracted from the syntactic parse trees of sentences. The importance of each word is inferred from graph connectivities. Second, the local syntactic parsing structure of words is considered in computing the distance between words. To demonstrate the effectiveness of the proposed SynWMD, we conduct experiments on 6 textual semantic similarity (STS) datasets and 4 sentence classification datasets. Experimental results show that SynWMD achieves state-of-the-art performance on STS tasks. It also outperforms other WMD-based methods on sentence classification tasks.

The design of robust learning systems that offer stable performance under a wide range of supervision degrees is investigated in this work. We choose the image classification problem as an illustrative example and focus on the design of modularized systems that consist of three learning modules: representation learning, feature learning and decision learning. We discuss ways to adjust each module so that the design is robust with respect to different training sample numbers. Based on these ideas, we propose two families of learning systems. One adopts the classical histogram of oriented gradients (HOG) features while the other uses successive-subspace-learning (SSL) features. We test their performance against LeNet-5, which is an end-to-end optimized neural network, for MNIST and Fashion-MNIST datasets. The number of training samples per image class goes from the extremely weak supervision condition (i.e., 1 labeled sample per class) to the strong supervision condition (i.e., 4096 labeled sample per class) with gradual transition in between (i.e., $2^n$, $n=0, 1, \cdots, 12$). Experimental results show that the two families of modularized learning systems have more robust performance than LeNet-5. They both outperform LeNet-5 by a large margin for small $n$ and have performance comparable with that of LeNet-5 for large $n$., Comment: 16 pages, 12 figures, 4 tables, under consideration at Pattern Recognition

Unsupervised domain adaptation (UDA) has been vastly explored to alleviate domain shifts between source and target domains, by applying a well-performed model in an unlabeled target domain via supervision of a labeled source domain. Recent literature, however, has indicated that the performance is still far from satisfactory in the presence of significant domain shifts. Nonetheless, delineating a few target samples is usually manageable and particularly worthwhile, due to the substantial performance gain. Inspired by this, we aim to develop semi-supervised domain adaptation (SSDA) for medical image segmentation, which is largely underexplored. We, thus, propose to exploit both labeled source and target domain data, in addition to unlabeled target data in a unified manner. Specifically, we present a novel asymmetric co-training (ACT) framework to integrate these subsets and avoid the domination of the source domain data. Following a divide-and-conquer strategy, we explicitly decouple the label supervisions in SSDA into two asymmetric sub-tasks, including semi-supervised learning (SSL) and UDA, and leverage different knowledge from two segmentors to take into account the distinction between the source and target label supervisions. The knowledge learned in the two modules is then adaptively integrated with ACT, by iteratively teaching each other, based on the confidence-aware pseudo-label. In addition, pseudo label noise is well-controlled with an exponential MixUp decay scheme for smooth propagation. Experiments on cross-modality brain tumor MRI segmentation tasks using the BraTS18 database showed, even with limited labeled target samples, ACT yielded marked improvements over UDA and state-of-the-art SSDA methods and approached an "upper bound" of supervised joint training., Comment: MICCAI 2022 (early accept)

This work presents a generative modeling approach based on successive subspace learning (SSL). Unlike most generative models in the literature, our method does not utilize neural networks to analyze the underlying source distribution and synthesize images. The resulting method, called the progressive attribute-guided extendable robust image generative (PAGER) model, has advantages in mathematical transparency, progressive content generation, lower training time, robust performance with fewer training samples, and extendibility to conditional image generation. PAGER consists of three modules: core generator, resolution enhancer, and quality booster. The core generator learns the distribution of low-resolution images and performs unconditional image generation. The resolution enhancer increases image resolution via conditional generation. Finally, the quality booster adds finer details to generated images. Extensive experiments on MNIST, Fashion-MNIST, and CelebA datasets are conducted to demonstrate generative performance of PAGER., Comment: 19 pages, 12 figures, 2 tables