Fan Zhu

Zhipeng Yao, Dazhou Li, Zitong Zhang et al.

Diffusion models have achieved remarkable success, yet their training remains inefficient due to a severe optimization bottleneck, which we term Representation Degradation. As noise levels increase, the outputs of the trained model exhibit progressive structural distortion, which can destabilize training and impair generation quality. Our analysis suggests that this instability is driven by mismatched target recoverability, which is associated with Neural Tangent Kernel (NTK) spectral weakening and effective low-rank behavior. To address this, we propose Elucidated Representation Diffusion (ERD), a plug-and-play framework that dynamically reallocates optimization effort according to effective recoverability. By stabilizing representation learning without external supervision, ERD accelerates convergence and achieves strong empirical performance across diffusion backbones.

Xin Xu, Yu Dong, Fan Zhu

Autonomous driving has been quite promising in recent years. The public has seen Robotaxi delivered by Waymo, Baidu, Cruise, and so on. While autonomous driving vehicles certainly have a bright future, we have to admit that it is still a long way to go for products such as Robotaxi. On the other hand, in less complex scenarios autonomous driving may have the potentiality to reliably outperform humans. For example, humans are good at interactive tasks (while autonomous driving systems usually do not), but we are often incompetent for tasks with strict precision demands. In this paper, we introduce a real-world, industrial scenario of which human drivers are not capable. The task required the ego vehicle to keep a stationary lateral distance (i.e. 3? <= 5 centimeters) with respect to a reference. To address this challenge, we redesigned the control module from Baidu Apollo open-source autonomous driving system. A precise (3? <= 2 centimeters) Error Feedback System was first built to partly replace the localization module. Then we investigated the control module thoroughly and added a real-time calibration algorithm to gain extra precision. We also built a simulation to fine-tune the control parameters. After all those works, the results are encouraging, showing that an end-to-end lateral precision with 3? <= 5 centimeters has been achieved. Further, we show that the results not only outperformed original Apollo modules but also beat specially trained and highly experienced human test drivers.

Yichao Yan, Jie Qin1, Jiaxin Chen et al.

Video-based person re-identification (re-ID) is an important research topic in computer vision. The key to tackling the challenging task is to exploit both spatial and temporal clues in video sequences. In this work, we propose a novel graph-based framework, namely Multi-Granular Hypergraph (MGH), to pursue better representational capabilities by modeling spatiotemporal dependencies in terms of multiple granularities. Specifically, hypergraphs with different spatial granularities are constructed using various levels of part-based features across the video sequence. In each hypergraph, different temporal granularities are captured by hyperedges that connect a set of graph nodes (i.e., part-based features) across different temporal ranges. Two critical issues (misalignment and occlusion) are explicitly addressed by the proposed hypergraph propagation and feature aggregation schemes. Finally, we further enhance the overall video representation by learning more diversified graph-level representations of multiple granularities based on mutual information minimization. Extensive experiments on three widely adopted benchmarks clearly demonstrate the effectiveness of the proposed framework. Notably, 90.0% top-1 accuracy on MARS is achieved using MGH, outperforming the state-of-the-arts. Code is available at https://github.com/daodaofr/hypergraph_reid.

Yichao Yan, Jie Qin, Bingbing Ni et al.

Learning to re-identify or retrieve a group of people across non-overlapped camera systems has important applications in video surveillance. However, most existing methods focus on (single) person re-identification (re-id), ignoring the fact that people often walk in groups in real scenarios. In this work, we take a step further and consider employing context information for identifying groups of people, i.e., group re-id. We propose a novel unified framework based on graph neural networks to simultaneously address the group-based re-id tasks, i.e., group re-id and group-aware person re-id. Specifically, we construct a context graph with group members as its nodes to exploit dependencies among different people. A multi-level attention mechanism is developed to formulate both intra-group and inter-group context, with an additional self-attention module for robust graph-level representations by attentively aggregating node-level features. The proposed model can be directly generalized to tackle group-aware person re-id using node-level representations. Meanwhile, to facilitate the deployment of deep learning models on these tasks, we build a new group re-id dataset that contains more than 3.8K images with 1.5K annotated groups, an order of magnitude larger than existing group re-id datasets. Extensive experiments on the novel dataset as well as three existing datasets clearly demonstrate the effectiveness of the proposed framework for both group-based re-id tasks. The code is available at https://github.com/daodaofr/group_reid.

Yichao Yan, Jinpeng Li, Jie Qin et al.

Person search aims to simultaneously localize and identify a query person from realistic, uncropped images, which can be regarded as the unified task of pedestrian detection and person re-identification (re-id). Most existing works employ two-stage detectors like Faster-RCNN, yielding encouraging accuracy but with high computational overhead. In this work, we present the Feature-Aligned Person Search Network (AlignPS), the first anchor-free framework to efficiently tackle this challenging task. AlignPS explicitly addresses the major challenges, which we summarize as the misalignment issues in different levels (i.e., scale, region, and task), when accommodating an anchor-free detector for this task. More specifically, we propose an aligned feature aggregation module to generate more discriminative and robust feature embeddings by following a "re-id first" principle. Such a simple design directly improves the baseline anchor-free model on CUHK-SYSU by more than 20% in mAP. Moreover, AlignPS outperforms state-of-the-art two-stage methods, with a higher speed. Code is available at https://github.com/daodaofr/AlignPS

Ionut Cosmin Duta, Li Liu, Fan Zhu et al.

This work introduces pyramidal convolution (PyConv), which is capable of processing the input at multiple filter scales. PyConv contains a pyramid of kernels, where each level involves different types of filters with varying size and depth, which are able to capture different levels of details in the scene. On top of these improved recognition capabilities, PyConv is also efficient and, with our formulation, it does not increase the computational cost and parameters compared to standard convolution. Moreover, it is very flexible and extensible, providing a large space of potential network architectures for different applications. PyConv has the potential to impact nearly every computer vision task and, in this work, we present different architectures based on PyConv for four main tasks on visual recognition: image classification, video action classification/recognition, object detection and semantic image segmentation/parsing. Our approach shows significant improvements over all these core tasks in comparison with the baselines. For instance, on image recognition, our 50-layers network outperforms in terms of recognition performance on ImageNet dataset its counterpart baseline ResNet with 152 layers, while having 2.39 times less parameters, 2.52 times lower computational complexity and more than 3 times less layers. On image segmentation, our novel framework sets a new state-of-the-art on the challenging ADE20K benchmark for scene parsing. Code is available at: https://github.com/iduta/pyconv

Ionut Cosmin Duta, Li Liu, Fan Zhu et al.

Residual networks (ResNets) represent a powerful type of convolutional neural network (CNN) architecture, widely adopted and used in various tasks. In this work we propose an improved version of ResNets. Our proposed improvements address all three main components of a ResNet: the flow of information through the network layers, the residual building block, and the projection shortcut. We are able to show consistent improvements in accuracy and learning convergence over the baseline. For instance, on ImageNet dataset, using the ResNet with 50 layers, for top-1 accuracy we can report a 1.19% improvement over the baseline in one setting and around 2% boost in another. Importantly, these improvements are obtained without increasing the model complexity. Our proposed approach allows us to train extremely deep networks, while the baseline shows severe optimization issues. We report results on three tasks over six datasets: image classification (ImageNet, CIFAR-10 and CIFAR-100), object detection (COCO) and video action recognition (Kinetics-400 and Something-Something-v2). In the deep learning era, we establish a new milestone for the depth of a CNN. We successfully train a 404-layer deep CNN on the ImageNet dataset and a 3002-layer network on CIFAR-10 and CIFAR-100, while the baseline is not able to converge at such extreme depths. Code is available at: https://github.com/iduta/iresnet

Yuming Shen, Jie Qin, Jiaxin Chen et al.

Conventional unsupervised hashing methods usually take advantage of similarity graphs, which are either pre-computed in the high-dimensional space or obtained from random anchor points. On the one hand, existing methods uncouple the procedures of hash function learning and graph construction. On the other hand, graphs empirically built upon original data could introduce biased prior knowledge of data relevance, leading to sub-optimal retrieval performance. In this paper, we tackle the above problems by proposing an efficient and adaptive code-driven graph, which is updated by decoding in the context of an auto-encoder. Specifically, we introduce into our framework twin bottlenecks (i.e., latent variables) that exchange crucial information collaboratively. One bottleneck (i.e., binary codes) conveys the high-level intrinsic data structure captured by the code-driven graph to the other (i.e., continuous variables for low-level detail information), which in turn propagates the updated network feedback for the encoder to learn more discriminative binary codes. The auto-encoding learning objective literally rewards the code-driven graph to learn an optimal encoder. Moreover, the proposed model can be simply optimized by gradient descent without violating the binary constraints. Experiments on benchmarked datasets clearly show the superiority of our framework over the state-of-the-art hashing methods. Our source code can be found at https://github.com/ymcidence/TBH.

Ziqin Wang, Jun Xu, Li Liu et al.

Despite online learning (OL) techniques have boosted the performance of semi-supervised video object segmentation (VOS) methods, the huge time costs of OL greatly restrict their practicality. Matching based and propagation based methods run at a faster speed by avoiding OL techniques. However, they are limited by sub-optimal accuracy, due to mismatching and drifting problems. In this paper, we develop a real-time yet very accurate Ranking Attention Network (RANet) for VOS. Specifically, to integrate the insights of matching based and propagation based methods, we employ an encoder-decoder framework to learn pixel-level similarity and segmentation in an end-to-end manner. To better utilize the similarity maps, we propose a novel ranking attention module, which automatically ranks and selects these maps for fine-grained VOS performance. Experiments on DAVIS-16 and DAVIS-17 datasets show that our RANet achieves the best speed-accuracy trade-off, e.g., with 33 milliseconds per frame and J&F=85.5% on DAVIS-16. With OL, our RANet reaches J&F=87.1% on DAVIS-16, exceeding state-of-the-art VOS methods. The code can be found at https://github.com/Storife/RANet.

Jun Xu, Yuan Huang, Ming-Ming Cheng et al.

Supervised deep networks have achieved promisingperformance on image denoising, by learning image priors andnoise statistics on plenty pairs of noisy and clean images. Unsupervised denoising networks are trained with only noisy images. However, for an unseen corrupted image, both supervised andunsupervised networks ignore either its particular image prior, the noise statistics, or both. That is, the networks learned from external images inherently suffer from a domain gap problem: the image priors and noise statistics are very different between the training and test images. This problem becomes more clear when dealing with the signal dependent realistic noise. To circumvent this problem, in this work, we propose a novel "Noisy-As-Clean" (NAC) strategy of training self-supervised denoising networks. Specifically, the corrupted test image is directly taken as the "clean" target, while the inputs are synthetic images consisted of this corrupted image and a second and similar corruption. A simple but useful observation on our NAC is: as long as the noise is weak, it is feasible to learn a self-supervised network only with the corrupted image, approximating the optimal parameters of a supervised network learned with pairs of noisy and clean images. Experiments on synthetic and realistic noise removal demonstrate that, the DnCNN and ResNet networks trained with our self-supervised NAC strategy achieve comparable or better performance than the original ones and previous supervised/unsupervised/self-supervised networks. The code is publicly available at https://github.com/csjunxu/Noisy-As-Clean.

Yingkun Hou, Jun Xu, Mingxia Liu et al.

Non-local self similarity (NSS) is a powerful prior of natural images for image denoising. Most of existing denoising methods employ similar patches, which is a patch-level NSS prior. In this paper, we take one step forward by introducing a pixel-level NSS prior, i.e., searching similar pixels across a non-local region. This is motivated by the fact that finding closely similar pixels is more feasible than similar patches in natural images, which can be used to enhance image denoising performance. With the introduced pixel-level NSS prior, we propose an accurate noise level estimation method, and then develop a blind image denoising method based on the lifting Haar transform and Wiener filtering techniques. Experiments on benchmark datasets demonstrate that, the proposed method achieves much better performance than previous non-deep methods, and is still competitive with existing state-of-the-art deep learning based methods on real-world image denoising. The code is publicly available at https://github.com/njusthyk1972/NLH.

Jun Xu, Yingkun Hou, Dongwei Ren et al.

Retinex theory is developed mainly to decompose an image into the illumination and reflectance components by analyzing local image derivatives. In this theory, larger derivatives are attributed to the changes in reflectance, while smaller derivatives are emerged in the smooth illumination. In this paper, we utilize exponentiated local derivatives (with an exponent γ) of an observed image to generate its structure map and texture map. The structure map is produced by been amplified with γ > 1, while the texture map is generated by been shrank with γ < 1. To this end, we design exponential filters for the local derivatives, and present their capability on extracting accurate structure and texture maps, influenced by the choices of exponents γ. The extracted structure and texture maps are employed to regularize the illumination and reflectance components in Retinex decomposition. A novel Structure and Texture Aware Retinex (STAR) model is further proposed for illumination and reflectance decomposition of a single image. We solve the STAR model by an alternating optimization algorithm. Each sub-problem is transformed into a vectorized least squares regression, with closed-form solutions. Comprehensive experiments on commonly tested datasets demonstrate that, the proposed STAR model produce better quantitative and qualitative performance than previous competing methods, on illumination and reflectance decomposition, low-light image enhancement, and color correction. The code is publicly available at https://github.com/csjunxu/STAR.

Haoyang Fan, Fan Zhu, Changchun Liu et al.

In this manuscript, we introduce a real-time motion planning system based on the Baidu Apollo (open source) autonomous driving platform. The developed system aims to address the industrial level-4 motion planning problem while considering safety, comfort and scalability. The system covers multilane and single-lane autonomous driving in a hierarchical manner: (1) The top layer of the system is a multilane strategy that handles lane-change scenarios by comparing lane-level trajectories computed in parallel. (2) Inside the lane-level trajectory generator, it iteratively solves path and speed optimization based on a Frenet frame. (3) For path and speed optimization, a combination of dynamic programming and spline-based quadratic programming is proposed to construct a scalable and easy-to-tune framework to handle traffic rules, obstacle decisions and smoothness simultaneously. The planner is scalable to both highway and lower-speed city driving scenarios. We also demonstrate the algorithm through scenario illustrations and on-road test results. The system described in this manuscript has been deployed to dozens of Baidu Apollo autonomous driving vehicles since Apollo v1.5 was announced in September 2017. As of May 16th, 2018, the system has been tested under 3,380 hours and approximately 68,000 kilometers (42,253 miles) of closed-loop autonomous driving under various urban scenarios. The algorithm described in this manuscript is available at https://github.com/ApolloAuto/apollo/tree/master/modules/planning.

Ziyu Chen, Fan Zhu, Hui Zhu et al.

Recent 3D Gaussian Splatting (3DGS) methods have demonstrated the feasibility of self-driving scene reconstruction and novel view synthesis. However, most existing methods either rely solely on cameras or use LiDAR only for Gaussian initialization or depth supervision, while the rich scene information contained in point clouds, such as reflectance, and the complementarity between LiDAR and RGB have not been fully exploited, leading to degradation in challenging self-driving scenes, such as those with high ego-motion and complex lighting. To address these issues, we propose a robust and efficient LiDAR-reflectance-guided Salient Gaussian Splatting method (LR-SGS) for self-driving scenes, which introduces a structure-aware Salient Gaussian representation, initialized from geometric and reflectance feature points extracted from LiDAR and refined through a salient transform and improved density control to capture edge and planar structures. Furthermore, we calibrate LiDAR intensity into reflectance and attach it to each Gaussian as a lighting-invariant material channel, jointly aligned with RGB to enforce boundary consistency. Extensive experiments on the Waymo Open Dataset demonstrate that LR-SGS achieves superior reconstruction performance with fewer Gaussians and shorter training time. In particular, on Complex Lighting scenes, our method surpasses OmniRe by 1.18 dB PSNR.

Fan Zhu, Ruixing Jia, Lei Yang et al.

We propose a deep visuo-tactile model for realtime estimation of the liquid inside a deformable container in a proprioceptive way.We fuse two sensory modalities, i.e., the raw visual inputs from the RGB camera and the tactile cues from our specific tactile sensor without any extra sensor calibrations.The robotic system is well controlled and adjusted based on the estimation model in real time. The main contributions and novelties of our work are listed as follows: 1) Explore a proprioceptive way for liquid volume estimation by developing an end-to-end predictive model with multi-modal convolutional networks, which achieve a high precision with an error of around 2 ml in the experimental validation. 2) Propose a multi-task learning architecture which comprehensively considers the losses from both classification and regression tasks, and comparatively evaluate the performance of each variant on the collected data and actual robotic platform. 3) Utilize the proprioceptive robotic system to accurately serve and control the requested volume of liquid, which is continuously flowing into a deformable container in real time. 4) Adaptively adjust the grasping plan to achieve more stable grasping and manipulation according to the real-time liquid volume prediction.

Hao Huang, Boulbaba Ben Amor, Xichan Lin et al.

Analyzing the structure of proteins is a key part of understanding their functions and thus their role in biology at the molecular level. In addition, design new proteins in a methodical way is a major engineering challenge. In this work, we introduce a joint geometric-neural networks approach for comparing, deforming and generating 3D protein structures. Viewing protein structures as 3D open curves, we adopt the Square Root Velocity Function (SRVF) representation and leverage its suitable geometric properties along with Deep Residual Networks (ResNets) for a joint registration and comparison. Our ResNets handle better large protein deformations while being more computationally efficient. On top of the mathematical framework, we further design a Geometric Variational Auto-Encoder (G-VAE), that once trained, maps original, previously unseen structures, into a low-dimensional (latent) hyper-sphere. Motivated by the spherical structure of the pre-shape space, we naturally adopt the von Mises-Fisher (vMF) distribution to model our hidden variables. We test the effectiveness of our models by generating novel protein structures and predicting completions of corrupted protein structures. Experimental results show that our method is able to generate plausible structures, different from the structures in the training data.

Hao Huang, Boulbaba Ben Amor, Xichan Lin et al.

Non-linear (large) time warping is a challenging source of nuisance in time-series analysis. In this paper, we propose a novel diffeomorphic temporal transformer network for both pairwise and joint time-series alignment. Our ResNet-TW (Deep Residual Network for Time Warping) tackles the alignment problem by compositing a flow of incremental diffeomorphic mappings. Governed by the flow equation, our Residual Network (ResNet) builds smooth, fluid and regular flows of velocity fields and consequently generates smooth and invertible transformations (i.e. diffeomorphic warping functions). Inspired by the elegant Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework, the final transformation is built by the flow of time-dependent vector fields which are none other than the building blocks of our Residual Network. The latter is naturally viewed as an Eulerian discretization schema of the flow equation (an ODE). Once trained, our ResNet-TW aligns unseen data by a single inexpensive forward pass. As we show in experiments on both univariate (84 datasets from UCR archive) and multivariate time-series (MSR Action-3D, Florence-3D and MSR Daily Activity), ResNet-TW achieves competitive performance in joint alignment and classification.

Bing Wang, Changhao Chen, Zhaopeng Cui et al.

Accurately describing and detecting 2D and 3D keypoints is crucial to establishing correspondences across images and point clouds. Despite a plethora of learning-based 2D or 3D local feature descriptors and detectors having been proposed, the derivation of a shared descriptor and joint keypoint detector that directly matches pixels and points remains under-explored by the community. This work takes the initiative to establish fine-grained correspondences between 2D images and 3D point clouds. In order to directly match pixels and points, a dual fully convolutional framework is presented that maps 2D and 3D inputs into a shared latent representation space to simultaneously describe and detect keypoints. Furthermore, an ultra-wide reception mechanism in combination with a novel loss function are designed to mitigate the intrinsic information variations between pixel and point local regions. Extensive experimental results demonstrate that our framework shows competitive performance in fine-grained matching between images and point clouds and achieves state-of-the-art results for the task of indoor visual localization. Our source code will be available at [no-name-for-blind-review].

Lei Huang, Yi Zhou, Li Liu et al.

Batch normalization (BN) is an important technique commonly incorporated into deep learning models to perform standardization within mini-batches. The merits of BN in improving a model's learning efficiency can be further amplified by applying whitening, while its drawbacks in estimating population statistics for inference can be avoided through group normalization (GN). This paper proposes group whitening (GW), which exploits the advantages of the whitening operation and avoids the disadvantages of normalization within mini-batches. In addition, we analyze the constraints imposed on features by normalization, and show how the batch size (group number) affects the performance of batch (group) normalized networks, from the perspective of model's representational capacity. This analysis provides theoretical guidance for applying GW in practice. Finally, we apply the proposed GW to ResNet and ResNeXt architectures and conduct experiments on the ImageNet and COCO benchmarks. Results show that GW consistently improves the performance of different architectures, with absolute gains of $1.02\%$ $\sim$ $1.49\%$ in top-1 accuracy on ImageNet and $1.82\%$ $\sim$ $3.21\%$ in bounding box AP on COCO.

Lei Huang, Jie Qin, Yi Zhou et al.

Normalization techniques are essential for accelerating the training and improving the generalization of deep neural networks (DNNs), and have successfully been used in various applications. This paper reviews and comments on the past, present and future of normalization methods in the context of DNN training. We provide a unified picture of the main motivation behind different approaches from the perspective of optimization, and present a taxonomy for understanding the similarities and differences between them. Specifically, we decompose the pipeline of the most representative normalizing activation methods into three components: the normalization area partitioning, normalization operation and normalization representation recovery. In doing so, we provide insight for designing new normalization technique. Finally, we discuss the current progress in understanding normalization methods, and provide a comprehensive review of the applications of normalization for particular tasks, in which it can effectively solve the key issues.

Lei Huang, Li Liu, Fan Zhu et al.

Orthogonality is widely used for training deep neural networks (DNNs) due to its ability to maintain all singular values of the Jacobian close to 1 and reduce redundancy in representation. This paper proposes a computationally efficient and numerically stable orthogonalization method using Newton's iteration (ONI), to learn a layer-wise orthogonal weight matrix in DNNs. ONI works by iteratively stretching the singular values of a weight matrix towards 1. This property enables it to control the orthogonality of a weight matrix by its number of iterations. We show that our method improves the performance of image classification networks by effectively controlling the orthogonality to provide an optimal tradeoff between optimization benefits and representational capacity reduction. We also show that ONI stabilizes the training of generative adversarial networks (GANs) by maintaining the Lipschitz continuity of a network, similar to spectral normalization (SN), and further outperforms SN by providing controllable orthogonality.

Lei Huang, Lei Zhao, Yi Zhou et al.

Batch Normalization (BN) is extensively employed in various network architectures by performing standardization within mini-batches. A full understanding of the process has been a central target in the deep learning communities. Unlike existing works, which usually only analyze the standardization operation, this paper investigates the more general Batch Whitening (BW). Our work originates from the observation that while various whitening transformations equivalently improve the conditioning, they show significantly different behaviors in discriminative scenarios and training Generative Adversarial Networks (GANs). We attribute this phenomenon to the stochasticity that BW introduces. We quantitatively investigate the stochasticity of different whitening transformations and show that it correlates well with the optimization behaviors during training. We also investigate how stochasticity relates to the estimation of population statistics during inference. Based on our analysis, we provide a framework for designing and comparing BW algorithms in different scenarios. Our proposed BW algorithm improves the residual networks by a significant margin on ImageNet classification. Besides, we show that the stochasticity of BW can improve the GAN's performance with, however, the sacrifice of the training stability.

Lei Huang, Jie Qin, Li Liu et al.

Conditioning analysis uncovers the landscape of an optimization objective by exploring the spectrum of its curvature matrix. This has been well explored theoretically for linear models. We extend this analysis to deep neural networks (DNNs) in order to investigate their learning dynamics. To this end, we propose layer-wise conditioning analysis, which explores the optimization landscape with respect to each layer independently. Such an analysis is theoretically supported under mild assumptions that approximately hold in practice. Based on our analysis, we show that batch normalization (BN) can stabilize the training, but sometimes result in the false impression of a local minimum, which has detrimental effects on the learning. Besides, we experimentally observe that BN can improve the layer-wise conditioning of the optimization problem. Finally, we find that the last linear layer of a very deep residual network displays ill-conditioned behavior. We solve this problem by only adding one BN layer before the last linear layer, which achieves improved performance over the original and pre-activation residual networks.

Huan Xiong, Mengyang Yu, Li Liu et al.

Binary optimization, a representative subclass of discrete optimization, plays an important role in mathematical optimization and has various applications in computer vision and machine learning. Usually, binary optimization problems are NP-hard and difficult to solve due to the binary constraints, especially when the number of variables is very large. Existing methods often suffer from high computational costs or large accumulated quantization errors, or are only designed for specific tasks. In this paper, we propose a fast algorithm to find effective approximate solutions for general binary optimization problems. The proposed algorithm iteratively solves minimization problems related to the linear surrogates of loss functions, which leads to the updating of some binary variables most impacting the value of loss functions in each step. Our method supports a wide class of empirical objective functions with/without restrictions on the numbers of $1$s and $-1$s in the binary variables. Furthermore, the theoretical convergence of our algorithm is proven, and the explicit convergence rates are derived, for objective functions with Lipschitz continuous gradients, which are commonly adopted in practice. Extensive experiments on several binary optimization tasks and large-scale datasets demonstrate the superiority of the proposed algorithm over several state-of-the-art methods in terms of both effectiveness and efficiency.

Yuming Shen, Jie Qin, Jiaxin Chen et al.

Recent binary representation learning models usually require sophisticated binary optimization, similarity measure or even generative models as auxiliaries. However, one may wonder whether these non-trivial components are needed to formulate practical and effective hashing models. In this paper, we answer the above question by proposing an embarrassingly simple approach to binary representation learning. With a simple classification objective, our model only incorporates two additional fully-connected layers onto the top of an arbitrary backbone network, whilst complying with the binary constraints during training. The proposed model lower-bounds the Information Bottleneck (IB) between data samples and their semantics, and can be related to many recent `learning to hash' paradigms. We show that, when properly designed, even such a simple network can generate effective binary codes, by fully exploring data semantics without any held-out alternating updating steps or auxiliary models. Experiments are conducted on conventional large-scale benchmarks, i.e., CIFAR-10, NUS-WIDE, and ImageNet, where the proposed simple model outperforms the state-of-the-art methods.

Hongyu Li, Fan Zhu, Junhua Qiu

Since the low quality of document images will greatly undermine the chances of success in automatic text recognition and analysis, it is necessary to assess the quality of document images uploaded in online business process, so as to reject those images of low quality. In this paper, we attempt to achieve document image quality assessment and our contributions are twofold. Firstly, since document image quality assessment is more interested in text, we propose a text line based framework to estimate document image quality, which is composed of three stages: text line detection, text line quality prediction, and overall quality assessment. Text line detection aims to find potential text lines with a detector. In the text line quality prediction stage, the quality score is computed for each text line with a CNN-based prediction model. The overall quality of document images is finally assessed with the ensemble of all text line quality. Secondly, to train the prediction model, a large-scale dataset, comprising 52,094 text line images, is synthesized with diverse attributes. For each text line image, a quality label is computed with a piece-wise function. To demonstrate the effectiveness of the proposed framework, comprehensive experiments are evaluated on two popular document image quality assessment benchmarks. Our framework significantly outperforms the state-of-the-art methods by large margins on the large and complicated dataset.

Syed Waqas Zamir, Aditya Arora, Akshita Gupta et al.

Existing Earth Vision datasets are either suitable for semantic segmentation or object detection. In this work, we introduce the first benchmark dataset for instance segmentation in aerial imagery that combines instance-level object detection and pixel-level segmentation tasks. In comparison to instance segmentation in natural scenes, aerial images present unique challenges e.g., a huge number of instances per image, large object-scale variations and abundant tiny objects. Our large-scale and densely annotated Instance Segmentation in Aerial Images Dataset (iSAID) comes with 655,451 object instances for 15 categories across 2,806 high-resolution images. Such precise per-pixel annotations for each instance ensure accurate localization that is essential for detailed scene analysis. Compared to existing small-scale aerial image based instance segmentation datasets, iSAID contains 15$\times$ the number of object categories and 5$\times$ the number of instances. We benchmark our dataset using two popular instance segmentation approaches for natural images, namely Mask R-CNN and PANet. In our experiments we show that direct application of off-the-shelf Mask R-CNN and PANet on aerial images provide suboptimal instance segmentation results, thus requiring specialized solutions from the research community. The dataset is publicly available at: https://captain-whu.github.io/iSAID/index.html

Yazhou Yao, Zeren Sun, Fumin Shen et al.

Due to the high cost of manual annotation, learning directly from the web has attracted broad attention. One issue that limits their performance is the problem of visual polysemy. To address this issue, we present an adaptive multi-model framework that resolves polysemy by visual disambiguation. Compared to existing methods, the primary advantage of our approach lies in that our approach can adapt to the dynamic changes in the search results. Our proposed framework consists of two major steps: we first discover and dynamically select the text queries according to the image search results, then we employ the proposed saliency-guided deep multi-instance learning network to remove outliers and learn classification models for visual disambiguation. Extensive experiments demonstrate the superiority of our proposed approach.

Lei Huang, Yi Zhou, Fan Zhu et al.

Batch Normalization (BN) is ubiquitously employed for accelerating neural network training and improving the generalization capability by performing standardization within mini-batches. Decorrelated Batch Normalization (DBN) further boosts the above effectiveness by whitening. However, DBN relies heavily on either a large batch size, or eigen-decomposition that suffers from poor efficiency on GPUs. We propose Iterative Normalization (IterNorm), which employs Newton's iterations for much more efficient whitening, while simultaneously avoiding the eigen-decomposition. Furthermore, we develop a comprehensive study to show IterNorm has better trade-off between optimization and generalization, with theoretical and experimental support. To this end, we exclusively introduce Stochastic Normalization Disturbance (SND), which measures the inherent stochastic uncertainty of samples when applied to normalization operations. With the support of SND, we provide natural explanations to several phenomena from the perspective of optimization, e.g., why group-wise whitening of DBN generally outperforms full-whitening and why the accuracy of BN degenerates with reduced batch sizes. We demonstrate the consistently improved performance of IterNorm with extensive experiments on CIFAR-10 and ImageNet over BN and DBN.

Zheng Zhang, Li Liu, Jie Qin et al.

How to economically cluster large-scale multi-view images is a long-standing problem in computer vision. To tackle this challenge, we introduce a novel approach named Highly-economized Scalable Image Clustering (HSIC) that radically surpasses conventional image clustering methods via binary compression. We intuitively unify the binary representation learning and efficient binary cluster structure learning into a joint framework. In particular, common binary representations are learned by exploiting both sharable and individual information across multiple views to capture their underlying correlations. Meanwhile, cluster assignment with robust binary centroids is also performed via effective discrete optimization under L21-norm constraint. By this means, heavy continuous-valued Euclidean distance computations can be successfully reduced by efficient binary XOR operations during the clustering procedure. To our best knowledge, HSIC is the first binary clustering work specifically designed for scalable multi-view image clustering. Extensive experimental results on four large-scale image datasets show that HSIC consistently outperforms the state-of-the-art approaches, whilst significantly reducing computational time and memory footprint.

Fan Zhu, Lin Ma, Xin Xu et al.

For any autonomous driving vehicle, control module determines its road performance and safety, i.e. its precision and stability should stay within a carefully-designed range. Nonetheless, control algorithms require vehicle dynamics (such as longitudinal dynamics) as inputs, which, unfortunately, are obscure to calibrate in real time. As a result, to achieve reasonable performance, most, if not all, research-oriented autonomous vehicles do manual calibrations in a one-by-one fashion. Since manual calibration is not sustainable once entering into mass production stage for industrial purposes, we here introduce a machine-learning based auto-calibration system for autonomous driving vehicles. In this paper, we will show how we build a data-driven longitudinal calibration procedure using machine learning techniques. We first generated offline calibration tables from human driving data. The offline table serves as an initial guess for later uses and it only needs twenty-minutes data collection and process. We then used an online-learning algorithm to appropriately update the initial table (the offline table) based on real-time performance analysis. This longitudinal auto-calibration system has been deployed to more than one hundred Baidu Apollo self-driving vehicles (including hybrid family vehicles and electronic delivery-only vehicles) since April 2018. By August 27, 2018, it had been tested for more than two thousands hours, ten thousands kilometers (6,213 miles) and yet proven to be effective.

Hongyu Li, Fan Zhu, Junhua Qiu

Document image quality assessment (DIQA) is an important and challenging problem in real applications. In order to predict the quality scores of document images, this paper proposes a novel no-reference DIQA method based on character gradient, where the OCR accuracy is used as a ground-truth quality metric. Character gradient is computed on character patches detected with the maximally stable extremal regions (MSER) based method. Character patches are essentially significant to character recognition and therefore suitable for use in estimating document image quality. Experiments on a benchmark dataset show that the proposed method outperforms the state-of-the-art methods in estimating the quality score of document images.

Yazhou Yao, Jian Zhang, Fumin Shen et al.

The availability of labeled image datasets has been shown critical for high-level image understanding, which continuously drives the progress of feature designing and models developing. However, constructing labeled image datasets is laborious and monotonous. To eliminate manual annotation, in this work, we propose a novel image dataset construction framework by employing multiple textual queries. We aim at collecting diverse and accurate images for given queries from the Web. Specifically, we formulate noisy textual queries removing and noisy images filtering as a multi-view and multi-instance learning problem separately. Our proposed approach not only improves the accuracy but also enhances the diversity of the selected images. To verify the effectiveness of our proposed approach, we construct an image dataset with 100 categories. The experiments show significant performance gains by using the generated data of our approach on several tasks, such as image classification, cross-dataset generalization, and object detection. The proposed method also consistently outperforms existing weakly supervised and web-supervised approaches.