Search

In this paper, we propose to train binarized convolutional neural networks (CNNs) that are of significant importance for deploying deep learning to mobile devices with limited power capacity and computing resources. Previous works on quantizing CNNs often seek to approximate the floating-point information of weights and/or activations using a set of discrete values. Such methods, termed value approximation here, typically are built on the same network architecture of the full-precision counterpart. Instead, we take a new "structured approximation" view for network quantization — it is possible and valuable to exploit flexible architecture transformation when learning low-bit networks, which can achieve even better performance than the original networks in some cases. In particular, we propose a "group decomposition" strategy, termed GroupNet, which divides a network into desired groups. Interestingly, with our GroupNet strategy, each full-precision group can be effectively reconstructed by aggregating a set of homogeneous binary branches. We also propose to learn effective connections among groups to improve the representation capability. To improve the model capacity, we propose to dynamically execute sparse binary branches conditioned on input features while preserving the computational cost. More importantly, the proposed GroupNet shows strong flexibility for a few vision tasks. For instance, we extend the GroupNet for accurate semantic segmentation by embedding the rich context into the binary structure. The proposed GroupNet also shows strong performance on object detection. Experiments on image classification, semantic segmentation, and object detection tasks demonstrate the superior performance of the proposed methods over various quantized networks in the literature. Moreover, the speedup and runtime memory cost evaluation comparing with related quantization strategies is analyzed on GPU platforms, which serves as a strong benchmark for further research. [ABSTRACT FROM AUTHOR]

Before deploying a monocular depth estimation (MDE) model in real-world applications such as autonomous driving, it is critical to understand its generalization and robustness. Although the generalization of MDE models has been thoroughly studied, the robustness of the models has been overlooked in previous research. Existing state-of-the-art methods exhibit strong generalization to clean, unseen scenes. Such methods, however, appear to degrade when the test image is perturbed. This is likely because the prior arts typically use the primary 2D data augmentations (e.g., random horizontal flipping, random cropping, and color jittering), ignoring other common image degradation or corruptions. To mitigate this issue, we delve deeper into data augmentation and propose utilizing strong data augmentation techniques for robust depth estimation. In particular, we introduce 3D-aware defocus blur in addition to seven 2D data augmentations. We evaluate the generalization of our model on six clean RGB-D datasets that were not seen during training. To evaluate the robustness of MDE models, we create a benchmark by applying 15 common corruptions to the clean images from IBIMS, NYUDv2, KITTI, ETH3D, DIODE, and TUM. On this benchmark, we systematically study the robustness of our method and 9 representative MDE models. The experimental results demonstrate that our model exhibits better generalization and robustness than the previous methods. Specifically, we provide valuable insights about the choices of data augmentation strategies and network architectures, which would be useful for future research in robust monocular depth estimation. Our code, model, and benchmark can be available at https://github.com/KexianHust/Robust-MonoDepth. [ABSTRACT FROM AUTHOR]

We propose a branch-and-cut (B&C) method for solving general MAP-MRF inference problems. The core of our method is a very efficient bounding procedure, which combines scalable semidefinite programming (SDP) and a cutting-plane method for seeking violated constraints. In order to further speed up the computation, several strategies have been exploited, including model reduction, warm start and removal of inactive constraints. We analyze the performance of the proposed method under different settings, and demonstrate that our method either outperforms or performs on par with state-of-the-art approaches. Especially when the connectivities are dense or when the relative magnitudes of the unary costs are low, we achieve the best reported results. Experiments show that the proposed algorithm achieves better approximation than the state-of-the-art methods within a variety of time budgets on challenging non-submodular MAP-MRF inference problems. [ABSTRACT FROM AUTHOR]

We propose a fast, accurate matching method for estimating dense pixel correspondences across scenes. It is a challenging problem to estimate dense pixel correspondences between images depicting different scenes or instances of the same object category. While most such matching methods rely on hand-crafted features such as SIFT, we learn features from a large amount of unlabeled image patches using unsupervised learning. Pixel-layer features are obtained by encoding over the dictionary, followed by spatial pooling to obtain patch-layer features. The learned features are then seamlessly embedded into a multi-layer matching framework. We experimentally demonstrate that the learned features, together with our matching model, outperform state-of-the-art methods such as the SIFT flow (Liu et al. in IEEE Trans Pattern Anal Mach Intell 33(5):978-994, ), coherency sensitive hashing (Korman and Avidan in: Proceedings of the IEEE international conference on computer vision (ICCV), ) and the recent deformable spatial pyramid matching (Kim et al. in: Proceedings of the IEEE conference on computer vision and pattern recognition, ) methods both in terms of accuracy and computation efficiency. Furthermore, we evaluate the performance of a few different dictionary learning and feature encoding methods in the proposed pixel correspondence estimation framework, and analyze the impact of dictionary learning and feature encoding with respect to the final matching performance. [ABSTRACT FROM AUTHOR]

Cascade classifiers are widely used in real-time object detection. Different from conventional classifiers that are designed for a low overall classification error rate, a classifier in each node of the cascade is required to achieve an extremely high detection rate and moderate false positive rate. Although there are a few reported methods addressing this requirement in the context of object detection, there is no principled feature selection method that explicitly takes into account this asymmetric node learning objective. We provide such an algorithm here. We show that a special case of the biased minimax probability machine has the same formulation as the linear asymmetric classifier (LAC) of Wu et al. (linear asymmetric classifier for cascade detectors, ). We then design a new boosting algorithm that directly optimizes the cost function of LAC. The resulting totally-corrective boosting algorithm is implemented by the column generation technique in convex optimization. Experimental results on object detection verify the effectiveness of the proposed boosting algorithm as a node classifier in cascade object detection, and show performance better than that of the current state-of-the-art. [ABSTRACT FROM AUTHOR]

The state-of-the-art semantic segmentation methods have achieved impressive performance on predefined close-set individual datasets, but their generalization to zero-shot domains and unseen categories is limited. Labeling a large-scale dataset is challenging and expensive, Training a robust semantic segmentation model on multi-domains has drawn much attention. However, inconsistent taxonomies hinder the naive merging of current publicly available annotations. To address this, we propose a simple solution to scale up the multi-domain semantic segmentation dataset with less human effort. We replace each class label with a sentence embedding, which is a vector-valued embedding of a sentence describing the class. This approach enables the merging of multiple datasets from different domains, each with varying class labels and semantics. We merged publicly available noisy and weak annotations with the most finely annotated data, over 2 million images, which enables training a model that achieves performance equal to that of state-of-the-art supervised methods on 7 benchmark datasets, despite not using any images therefrom. Instead of manually tuning a consistent label space, we utilized a vector-valued embedding of short paragraphs to describe the classes. By fine-tuning the model on standard semantic segmentation datasets, we also achieve a significant improvement over the state-of-the-art supervised segmentation on NYUD-V2 (Silberman et al., in: European conference on computer vision, Springer, pp 746–760, 2012) and PASCAL-context (Everingham et al. in Int J Comput Visi 111(1):98–136, 2015) at 60%\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$60\%$$\end{document} and 65%\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$65\%$$\end{document} mIoU, respectively. Our method can segment unseen labels based on the closeness of language embeddings, showing strong generalization to unseen image domains and labels. Additionally, it enables impressive performance improvements in some adaptation applications, such as depth estimation and instance segmentation. Code is available at https://github.com/YvanYin/SSIW. [ABSTRACT FROM AUTHOR]

This paper investigates the capability of plain Vision Transformers (ViTs) for semantic segmentation using the encoder–decoder framework and introduce SegViTv2. In this study, we introduce a novel Attention-to-Mask (ATM) module to design a lightweight decoder effective for plain ViT. The proposed ATM converts the global attention map into semantic masks for high-quality segmentation results. Our decoder outperforms popular decoder UPerNet using various ViT backbones while consuming only about 5 % of the computational cost. For the encoder, we address the concern of the relatively high computational cost in the ViT-based encoders and propose a Shrunk++ structure that incorporates edge-aware query-based down-sampling (EQD) and query-based up-sampling (QU) modules. The Shrunk++ structure reduces the computational cost of the encoder by up to 50 % while maintaining competitive performance. Furthermore, we propose to adapt SegViT for continual semantic segmentation, demonstrating nearly zero forgetting of previously learned knowledge. Experiments show that our proposed SegViTv2 surpasses recent segmentation methods on three popular benchmarks including ADE20k, COCO-Stuff-10k and PASCAL-Context datasets. The code is available through the following link: https://github.com/zbwxp/SegVit. [ABSTRACT FROM AUTHOR]

We attempt to reduce the computational costs in vision transformers (ViTs), which increase quadratically in the token number. We present a novel training paradigm that trains only one ViT model at a time, but is capable of providing improved image recognition performance with various computational costs. Here, the trained ViT model, termed super vision transformer (SuperViT), is empowered with the versatile ability to solve incoming patches of multiple sizes as well as preserve informative tokens with multiple keeping rates (the ratio of keeping tokens) to achieve good hardware efficiency for inference, given that the available hardware resources often change from time to time. Experimental results on ImageNet demonstrate that our SuperViT can considerably reduce the computational costs of ViT models with even performance increase. For example, we reduce 2 × FLOPs of DeiT-S while increasing the Top-1 accuracy by 0.2% and 0.7% for 1.5 × reduction. Also, our SuperViT significantly outperforms existing studies on efficient vision transformers. For example, when consuming the same amount of FLOPs, our SuperViT surpasses the recent state-of-the-art EViT by 1.1% when using DeiT-S as their backbones. The project of this work is made publicly available at https://github.com/lmbxmu/SuperViT. [ABSTRACT FROM AUTHOR]

Multi-task visual perception has a wide range of applications in scene understanding such as autonomous driving. In this work, we devise an efficient unified framework to solve multiple common perception tasks, including instance segmentation, semantic segmentation, monocular 3D detection, and depth estimation. Simply sharing the same visual feature representations for these tasks impairs the performance of tasks, while independent task-specific feature extractors lead to parameter redundancy and latency. Thus, we design two feature-merge branches to learn feature basis, which can be useful to, and thus shared by, multiple perception tasks. Then, each task takes the corresponding feature basis as the input of the prediction task head to fulfill a specific task. In particular, one feature merge branch is designed for instance-level recognition the other for dense predictions. To enhance inter-branch communication, the instance branch passes pixel-wise spatial information of each instance to the dense branch using efficient dynamic convolution weighting. Moreover, a simple but effective dynamic routing mechanism is proposed to isolate task-specific features and leverage common properties among tasks. Our proposed framework, termed D2BNet, demonstrates a unique approach to parameter-efficient predictions for multi-task perception. In addition, as tasks benefit from co-training with each other, our solution achieves on par results on partially labeled settings on nuScenes and outperforms previous works for 3D detection and depth estimation on the Cityscapes dataset with full supervision. [ABSTRACT FROM AUTHOR]

Deep neural networks have achieved remarkable success in single image super-resolution (SISR). The computing and memory requirements of these methods have hindered their application to broad classes of real devices with limited computing power, however. One approach to this problem has been lightweight network architectures that balance the super-resolution performance and the computation burden. In this study, we revisit this problem from an orthogonal view, and propose a novel learning strategy to maximize the pixel-wise fitting ability of a given lightweight network architecture. Considering that the initial performance of the lightweight network is very limited, we present an adaptive importance learning scheme for SISR that trains the network with an easy-to-complex paradigm by dynamically updating the importance of image pixels on the basis of the training loss. Specifically, we formulate the network training and the importance learning into a joint optimization problem. With a carefully designed importance penalty function, the importance of individual pixels can be gradually increased through solving a convex optimization problem. The training process thus begins with pixels that are easy to reconstruct, and gradually proceeds to more complex pixels as fitting improves. Furthermore, the proposed learning scheme is able to seamlessly assimilate knowledge from a more powerful teacher network in the form of importance initialization, thus obtaining better initial performance for the network. Through learning the network parameters, and updating pixel importance, the proposed learning scheme enables smaller, lightweight, networks to achieve better performance than has previously been possible. Extensive experiments on four benchmark datasets demonstrate the potential benefits of the proposed learning strategy in lightweight SISR network enhancement. In some cases, our learned network with only 25 % of the parameters and computational complexity can produce comparable or even better results than the corresponding full-parameter network. [ABSTRACT FROM AUTHOR]

Existing deep learning-based facial attribute recognition (FAR) methods rely heavily on large-scale labeled training data. Unfortunately, in many real-world applications, only limited labeled data are available, resulting in the performance deterioration of these methods. To address this issue, we propose a novel spatial-semantic patch learning network (SPL-Net), consisting of a multi-branch shared subnetwork (MSS), three auxiliary task subnetworks (ATS), and an FAR subnetwork, for attribute classification with limited labeled data. Considering the diversity of facial attributes, MSS includes a task-shared branch and four region branches, each of which contains cascaded dual cross attention modules to extract region-specific features. SPL-Net involves a two-stage learning procedure. In the first stage, MSS and ATS are jointly trained to perform three auxiliary tasks (i.e., a patch rotation task (PRT), a patch segmentation task (PST), and a patch classification task (PCT)), which exploit the spatial-semantic relationship on large-scale unlabeled facial data from various perspectives. Specifically, PRT encodes the spatial information of facial images based on self-supervised learning. PST and PCT respectively capture the pixel-level and image-level semantic information of facial images by leveraging a facial parsing model. Thus, a well-pretrained MSS is obtained. In the second stage, based on the pre-trained MSS, an FAR model is easily fine-tuned to predict facial attributes by requiring only a small amount of labeled data. Experimental results on challenging facial attribute datasets (including CelebA, LFWA, and MAAD) show the superiority of SPL-Net over several state-of-the-art methods in the case of limited labeled data. [ABSTRACT FROM AUTHOR]

Visual counting, a task that aims to estimate the number of objects from an image/video, is an open-set problem by nature as the number of population can vary in [ 0 , + ∞) in theory. However, collected data are limited in reality, which means that only a closed set is observed. Existing methods typically model this task through regression, while they are prone to suffer from unseen scenes with counts out of the scope of the closed set. In fact, counting has an interesting and exclusive property—spatially decomposable. A dense region can always be divided until sub-region counts are within the previously observed closed set. We therefore introduce the idea of spatial divide-and-conquer (S-DC) that transforms open-set counting into a closed set problem. This idea is implemented by a novel Supervised Spatial Divide-and-Conquer Network (SS-DCNet). It can learn from a closed set but generalize to open-set scenarios via S-DC. We provide mathematical analyses and a controlled experiment on synthetic data, demonstrating why closed-set modeling works well. Experiments show that SS-DCNet achieves state-of-the-art performance in crowd counting, vehicle counting and plant counting. SS-DCNet also demonstrates superior transferablity under the cross-dataset setting. Code and models are available at: https://git.io/SS-DCNet. [ABSTRACT FROM AUTHOR]

Hashing methods aim to learn a set of hash functions which map the original features to compact binary codes with similarity preserving in the Hamming space. Hashing has proven a valuable tool for large-scale information retrieval. We propose a column generation based binary code learning framework for data-dependent hash function learning. Given a set of triplets that encode the pairwise similarity comparison information, our column generation based method learns hash functions that preserve the relative comparison relations within the large-margin learning framework. Our method iteratively learns the best hash functions during the column generation procedure. Existing hashing methods optimize over simple objectives such as the reconstruction error or graph Laplacian related loss functions, instead of the performance evaluation criteria of interest-multivariate performance measures such as the AUC and NDCG. Our column generation based method can be further generalized from the triplet loss to a general structured learning based framework that allows one to directly optimize multivariate performance measures. For optimizing general ranking measures, the resulting optimization problem can involve exponentially or infinitely many variables and constraints, which is more challenging than standard structured output learning. We use a combination of column generation and cutting-plane techniques to solve the optimization problem. To speed-up the training we further explore stage-wise training and propose to optimize a simplified NDCG loss for efficient inference. We demonstrate the generality of our method by applying it to ranking prediction and image retrieval, and show that it outperforms several state-of-the-art hashing methods. [ABSTRACT FROM AUTHOR]

The purpose of mid-level visual element discovery is to find clusters of image patches that are representative of, and which discriminate between, the contents of the relevant images. Here we propose a pattern-mining approach to the problem of identifying mid-level elements within images, motivated by the observation that such techniques have been very effective, and efficient, in achieving similar goals when applied to other data types. We show that Convolutional Neural Network (CNN) activations extracted from image patches typical possess two appealing properties that enable seamless integration with pattern mining techniques. The marriage between CNN activations and a pattern mining technique leads to fast and effective discovery of representative and discriminative patterns from a huge number of image patches, from which mid-level elements are retrieved. Given the patterns and retrieved mid-level visual elements, we propose two methods to generate image feature representations. The first encoding method uses the patterns as codewords in a dictionary in a manner similar to the Bag-of-Visual-Words model. We thus label this a Bag-of-Patterns representation. The second relies on mid-level visual elements to construct a Bag-of-Elements representation. We evaluate the two encoding methods on object and scene classification tasks, and demonstrate that our approach outperforms or matches the performance of the state-of-the-arts on these tasks. [ABSTRACT FROM AUTHOR]

Sparsity-based representations have recently led to notable results in various visual recognition tasks. In a separate line of research, Riemannian manifolds have been shown useful for dealing with features and models that do not lie in Euclidean spaces. With the aim of building a bridge between the two realms, we address the problem of sparse coding and dictionary learning in Grassmann manifolds, i.e., the space of linear subspaces. To this end, we propose to embed Grassmann manifolds into the space of symmetric matrices by an isometric mapping. This in turn enables us to extend two sparse coding schemes to Grassmann manifolds. Furthermore, we propose an algorithm for learning a Grassmann dictionary, atom by atom. Lastly, to handle non-linearity in data, we extend the proposed Grassmann sparse coding and dictionary learning algorithms through embedding into higher dimensional Hilbert spaces. Experiments on several classification tasks (gender recognition, gesture classification, scene analysis, face recognition, action recognition and dynamic texture classification) show that the proposed approaches achieve considerable improvements in discrimination accuracy, in comparison to state-of-the-art methods such as kernelized Affine Hull Method and graph-embedding Grassmann discriminant analysis. [ABSTRACT FROM AUTHOR]

Visual tracking of generic objects is one of the fundamental but challenging problems in computer vision. Here, we propose a novel fully convolutional Siamese network to solve visual tracking by directly predicting the target bounding box in an end-to-end manner. We first reformulate the visual tracking task as two subproblems: a classification problem for pixel category prediction and a regression task for object status estimation at this pixel. With this decomposition, we design a simple yet effective Siamese architecture based classification and regression framework, termed SiamCAR, which consists of two subnetworks: a Siamese subnetwork for feature extraction and a classification-regression subnetwork for direct bounding box prediction. Since the proposed framework is both proposal- and anchor-free, SiamCAR can avoid the tedious hyper-parameter tuning of anchors, considerably simplifying the training. To demonstrate that a much simpler tracking framework can achieve superior tracking results, we conduct extensive experiments and comparisons with state-of-the-art trackers on a few challenging benchmarks. Without bells and whistles, SiamCAR achieves leading performance with a real-time speed. Furthermore, the ablation study validates that the proposed framework is effective with various backbone networks, and can benefit from deeper networks. Code is available at https://github.com/ohhhyeahhh/SiamCAR. [ABSTRACT FROM AUTHOR]

Recently, much attention has been spent on neural architecture search (NAS), aiming to outperform those manually-designed neural architectures on high-level vision recognition tasks. Inspired by the success, here we attempt to leverage NAS techniques to automatically design efficient network architectures for low-level image restoration tasks. In particular, we propose a memory-efficient hierarchical NAS (termed HiNAS) and apply it to two such tasks: image denoising and image super-resolution. HiNAS adopts gradient based search strategies and builds a flexible hierarchical search space, including the inner search space and outer search space. They are in charge of designing cell architectures and deciding cell widths, respectively. For the inner search space, we propose a layer-wise architecture sharing strategy, resulting in more flexible architectures and better performance. For the outer search space, we design a cell-sharing strategy to save memory, and considerably accelerate the search speed. The proposed HiNAS method is both memory and computation efficient. With a single GTX1080Ti GPU, it takes only about 1 h for searching for denoising network on the BSD-500 dataset and 3.5 h for searching for the super-resolution structure on the DIV2K dataset. Experiments show that the architectures found by HiNAS have fewer parameters and enjoy a faster inference speed, while achieving highly competitive performance compared with state-of-the-art methods. Code is available at: https://github.com/hkzhang91/HiNAS [ABSTRACT FROM AUTHOR]

Ghosting artifacts caused by moving objects and misalignments are a key challenge in constructing high dynamic range (HDR) images. Current methods first register the input low dynamic range (LDR) images using optical flow before merging them. This process is error-prone, and often causes ghosting in the resulting merged image. We propose a novel dual-attention-guided end-to-end deep neural network, called DAHDRNet, which produces high-quality ghost-free HDR images. Unlike previous methods that directly stack the LDR images or features for merging, we use dual-attention modules to guide the merging according to the reference image. DAHDRNet thus exploits both spatial attention and feature channel attention to achieve ghost-free merging. The spatial attention modules automatically suppress undesired components caused by misalignments and saturation, and enhance the fine details in the non-reference images. The channel attention modules adaptively rescale channel-wise features by considering the inter-dependencies between channels. The dual-attention approach is applied recurrently to further improve feature representation, and thus alignment. A dilated residual dense block is devised to make full use of the hierarchical features and increase the receptive field when hallucinating missing details. We employ a hybrid loss function, which consists of a perceptual loss, a total variation loss, and a content loss to recover photo-realistic images. Although DAHDRNet is not flow-based, it can be applied to flow-based registration to reduce artifacts caused by optical-flow estimation errors. Experiments on different datasets show that the proposed DAHDRNet achieves state-of-the-art quantitative and qualitative results. [ABSTRACT FROM AUTHOR]

Neural Architecture Search (NAS) has shown great potential in effectively reducing manual effort in network design by automatically discovering optimal architectures. What is noteworthy is that as of now, object detection is less touched by NAS algorithms despite its significant importance in computer vision. To the best of our knowledge, most of the recent NAS studies on object detection tasks fail to satisfactorily strike a balance between performance and efficiency of the resulting models, let alone the excessive amount of computational resources cost by those algorithms. Here we propose an efficient method to obtain better object detectors by searching for the feature pyramid network as well as the prediction head of a simple anchor-free object detector, namely, FCOS (Tian et al. in FCOS: Fully convolutional one-stage object detection, 2019), using a tailored reinforcement learning paradigm. With carefully designed search space, search algorithms, and strategies for evaluating network quality, we are able to find top-performing detection architectures within 4 days using 8 V100 GPUs. The discovered architectures surpass state-of-the-art object detection models (such as Faster R-CNN, RetinaNet and, FCOS) by 1.0 to 5.4% points in AP on the COCO dataset, with comparable computation complexity and memory footprint, demonstrating the efficacy of the proposed NAS method for object detection. Code is available at https://github.com/Lausannen/NAS-FCOS. [ABSTRACT FROM AUTHOR]

Low-level details and high-level semantics are both essential to the semantic segmentation task. However, to speed up the model inference, current approaches almost always sacrifice the low-level details, leading to a considerable decrease in accuracy. We propose to treat these spatial details and categorical semantics separately to achieve high accuracy and high efficiency for real-time semantic segmentation. For this purpose, we propose an efficient and effective architecture with a good trade-off between speed and accuracy, termed Bilateral Segmentation Network (BiSeNet V2). This architecture involves the following: (i) A detail branch, with wide channels and shallow layers to capture low-level details and generate high-resolution feature representation; (ii) A semantics branch, with narrow channels and deep layers to obtain high-level semantic context. The detail branch has wide channel dimensions and shallow layers, while the semantics branch has narrow channel dimensions and deep layers. Due to the reduction in the channel capacity and the use of a fast-downsampling strategy, the semantics branch is lightweight and can be implemented by any efficient model. We design a guided aggregation layer to enhance mutual connections and fuse both types of feature representation. Moreover, a booster training strategy is designed to improve the segmentation performance without any extra inference cost. Extensive quantitative and qualitative evaluations demonstrate that the proposed architecture shows favorable performance compared to several state-of-the-art real-time semantic segmentation approaches. Specifically, for a 2048 × 1024 input, we achieve 72.6% Mean IoU on the Cityscapes test set with a speed of 156 FPS on one NVIDIA GeForce GTX 1080 Ti card, which is significantly faster than existing methods, yet we achieve better segmentation accuracy. The code and trained models are available online at https://git.io/BiSeNet. [ABSTRACT FROM AUTHOR]

We propose a monocular depth estimation method SC-Depth, which requires only unlabelled videos for training and enables the scale-consistent prediction at inference time. Our contributions include: (i) we propose a geometry consistency loss, which penalizes the inconsistency of predicted depths between adjacent views; (ii) we propose a self-discovered mask to automatically localize moving objects that violate the underlying static scene assumption and cause noisy signals during training; (iii) we demonstrate the efficacy of each component with a detailed ablation study and show high-quality depth estimation results in both KITTI and NYUv2 datasets. Moreover, thanks to the capability of scale-consistent prediction, we show that our monocular-trained deep networks are readily integrated into ORB-SLAM2 system for more robust and accurate tracking. The proposed hybrid Pseudo-RGBD SLAM shows compelling results in KITTI, and it generalizes well to the KAIST dataset without additional training. Finally, we provide several demos for qualitative evaluation. The source code is released on GitHub. [ABSTRACT FROM AUTHOR]

Multi-orientation scene text detection has recently gained significant research attention. Previous methods directly predict words or text lines, typically by using quadrilateral shapes. However, many of these methods neglect the significance of consistent labeling, which is important for maintaining a stable training process, especially when it comprises a large amount of data. Here we solve this problem by proposing a new method, Orderless Box Discretization (OBD), which first discretizes the quadrilateral box into several key edges containing all potential horizontal and vertical positions. To decode accurate vertex positions, a simple yet effective matching procedure is proposed for reconstructing the quadrilateral bounding boxes. Our method solves the ambiguity issue, which has a significant impact on the learning process. Extensive ablation studies are conducted to validate the effectiveness of our proposed method quantitatively. More importantly, based on OBD, we provide a detailed analysis of the impact of a collection of refinements, which may inspire others to build state-of-the-art text detectors. Combining both OBD and these useful refinements, we achieve state-of-the-art performance on various benchmarks, including ICDAR 2015 and MLT. Our method also won the first place in the text detection task at the recent ICDAR2019 Robust Reading Challenge for Reading Chinese Text on Signboards, further demonstrating its superior performance. The code is available at https://git.io/TextDet. [ABSTRACT FROM AUTHOR]

Scene text recognition is an important task in computer vision. Despite tremendous progress achieved in the past few years, issues such as varying font styles, arbitrary shapes and complex backgrounds etc. have made the problem very challenging. In this work, we propose to improve text recognition from a new perspective by separating the text content from complex backgrounds, thus making the recognition considerably easier and significantly improving recognition accuracy. To this end, we exploit the generative adversarial networks (GANs) for removing backgrounds while retaining the text content. As vanilla GANs are not sufficiently robust to generate sequence-like characters in natural images, we propose an adversarial learning framework for the generation and recognition of multiple characters in an image. The proposed framework consists of an attention-based recognizer and a generative adversarial architecture. Furthermore, to tackle the issue of lacking paired training samples, we design an interactive joint training scheme, which shares attention masks from the recognizer to the discriminator, and enables the discriminator to extract the features of each character for further adversarial training. Benefiting from the character-level adversarial training, our framework requires only unpaired simple data for style supervision. Each target style sample containing only one randomly chosen character can be simply synthesized online during the training. This is significant as the training does not require costly paired samples or character-level annotations. Thus, only the input images and corresponding text labels are needed. In addition to the style normalization of the backgrounds, we refine character patterns to ease the recognition task. A feedback mechanism is proposed to bridge the gap between the discriminator and the recognizer. Therefore, the discriminator can guide the generator according to the confusion of the recognizer, so that the generated patterns are clearer for recognition. Experiments on various benchmarks, including both regular and irregular text, demonstrate that our method significantly reduces the difficulty of recognition. Our framework can be integrated into recent recognition methods to achieve new state-of-the-art recognition accuracy. [ABSTRACT FROM AUTHOR]

Hyperspectral images (HSIs) have significant advantages over more traditional image types for a variety of computer vision applications dues to the extra information available. The practical reality of capturing and transmitting HSIs however, means that they often exhibit large amounts of noise, or are undersampled to reduce the data volume. Methods for combating such image corruption are thus critical to many HSIs applications. Here we devise a novel cluster sparsity field (CSF) based HSI reconstruction framework which explicitly models both the intrinsic correlation between measurements within the spectrum for a particular pixel, and the similarity between pixels due to the spatial structure of the HSI. These two priors have been shown to be effective previously, but have been always considered separately. By dividing pixels of the HSI into a group of spatial clusters on the basis of spectrum characteristics, we define CSF, a Markov random field based prior. In CSF, a structured sparsity potential models the correlation between measurements within each spectrum, and a graph structure potential models the similarity between pixels in each spatial cluster. Then, we integrate the CSF prior learning and image reconstruction into a unified variational framework for optimization, which makes the CSF prior image-specific, and robust to noise. It also results in more accurate image reconstruction compared with existing HSI reconstruction methods, thus combating the effects of noise corruption or undersampling. Extensive experiments on HSI denoising and HSI compressive sensing demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

CaO-MgO-AlO-SiO (CMAS) glass was prepared by float process. The effects of TiO and heat-treatment on properties and crystallization behaviors of float glasses were investigated by atomic force microscope, differential scanning calorimeter, X-ray diffraction, electron probe microanalyzer, field emission scanning electron microscope and viscosity test. The results showed that CMAS parent glasses produced by float process had a high surface flatness ( Ra is less than 80.1 ± 0.1 nm) and low tin penetration (14 μm). When the concentration of TiO increased from 3.51 to 5.01 wt %, the glass transition temperature was decreased, and the crystallization temperature was shifted from 913 to 887°C using differential scanning calorimeter. Field emission scanning electron microscope images showed that phase separation was discovered in CMAS parent glass (containing 3.51 wt % TiO) treated at 670°C. Diopside as a major crystalline phase was precipitated in CMAS glass-ceramics nucleated at 700°C for 30 min and followed by crystallization at 910°C for 30 min. [ABSTRACT FROM AUTHOR]

It is difficult to research on the surface structure of amorphous phase in fly ash during leaching reaction due to crystalline phase and complex structure. In the present work, in order to reveal the effects of leaching reaction on the surface structure of amorphous phase in fly ash, the modelling CaO-FeO-AlO-SiO glass was prepared by the traditional melting methods. The leaching reaction of CaO-FeO-AlO-SiO glass with 7.5 M KOH was investigated by spectroscopy, spectrophotometer and wet chemical method. The results show that the content of Q, Q, Q and Q of glass without corrosion was 4.21, 9.51, 23.03 and 52.55%, respectively, which shows that the network polymerization of glass is compact. The leaching reaction of glass can be described by the following equation: d S/d t = k/( r + S). Leaching in KOH for various times induces the content of Q and Q to be decreased, and Q and Q increased, resulted in the depolymerization of network and the surface glass dissolved in alkaline solution to form a gel phase. In stage one of leaching reaction, the rate of iron ion leached from glass surface was slow, which resulted in the small slope of straight-line relationship of leaching curve. In the following stage, the leaching rate of iron ion increased with the prolongation of time. [ABSTRACT FROM AUTHOR]